Operation analysis method and device for wind turbine generator

Abstract

The invention discloses an operation analysis method and device for a wind turbine generator set, and relates to the field of electric digital data processing, the method analyzes the operation state of a wind turbine generator by using the characteristic that the vibration and the generating capacity of the wind turbine generator have a linear relation, and the relation between the generating capacity and the vibration is solved through linear regression. And meanwhile, a generating capacity-vibration prediction function is realized through linear regression, generating capacity of a wind turbine is calculated by considering wind speed probability distribution of the wind power plant, the generating capacity is substituted into a linear relation as a dependent variable for inversion, and vibration which should be generated is obtained. And finally, data obtained through forward deduction and data obtained through reverse deduction are mutually checked to obtain outlier data, the outlier is abnormal, a component corresponding to the abnormal position is marked, and the abnormal operation state of the wind turbine is analyzed.

Description

Technical Field

[0001] The present application relates to the technical field of electrical digital data processing, and in particular to an operation analysis method and device for a wind turbine generator set. Background Art

[0002] Wind power generation refers to the conversion of wind kinetic energy into electrical energy. Wind energy is a clean, pollution-free, renewable energy source. Wind power generation uses wind to drive the windmill blades to rotate, and then increases the speed of rotation through a speed increaser to drive the generator to generate electricity. Wind power generation does not require the use of fuel, and does not produce radiation or air pollution. It is a renewable new energy source.

[0003] Since wind farms are located in remote areas, wind turbines are operated outdoors for a long time, which makes the complex and changeable environmental conditions a challenge to the safe and stable operation of wind turbines. The design life of wind turbines is generally about 20 years, but the service life is often shortened due to damage to some key components. The early abnormal state of wind turbines is often manifested through vibration. In the process of wind turbines converting wind power mechanical energy into electrical energy, many parts of the wind turbines will vibrate. When this vibration reaches a certain level or exceeds a certain vibration amplitude, it will cause a fault, thereby damaging a component or even the entire wind turbine.

[0004] As wind farms are continuously built and put into operation, the number of wind turbines is increasing rapidly. Timely and effective operation, maintenance and repair of wind turbines is an urgent problem to be solved. Due to the limited number of wind turbine operation, maintenance and repair personnel and the single technical means of wind turbine operation status monitoring, the risk of wind turbine failure caused by the inability to timely maintain and repair wind turbines increases. Usually, wind farms will install an alarm system based on data acquisition and monitoring control systems to monitor the operating status of wind turbines, but because it is a post-fault alarm, maintenance personnel can only repair wind turbines after the fault occurs, and cannot provide maintenance personnel with wind turbine operating status information. For most wind farms, their maintenance and repair of wind turbines still mostly adopts regular inspections and post-fault repairs to maintain and repair wind turbines. Due to the shortage of wind turbine maintenance resources and limited technical conditions, it is difficult to achieve timely maintenance of wind turbines, resulting in the inability to carry out timely maintenance of wind turbines during inspection intervals. During the inspection interval, the wind turbine will only be repaired when a failure occurs, which not only causes maintenance downtime losses, but also increases the maintenance costs of wind turbines. Summary of the invention

[0005] The main purpose of the present application is to provide an operation analysis method, device, equipment and storage medium for a wind turbine generator set, so as to solve the problem in the prior art that the operating status of a wind turbine is monitored by an alarm system based on a data acquisition and monitoring control system, and an alarm is issued after a fault occurs. Maintenance personnel can only repair the wind turbine after the fault occurs, and it is impossible to provide maintenance personnel with information on the operating status of the wind turbine.

[0006] In order to achieve the above objectives, this application provides the following technical solutions: A method for analyzing the operation of a wind turbine generator set, wherein the wind turbine generator set includes a plurality of wind turbines located in the same wind farm area, and a vibration detection element is installed on each component of each wind turbine, and the method for analyzing the operation includes: Step S1, based on the current wind turbine, obtaining a number of actual vibration data of each component based on a number of preset natural time periods through all vibration detection components; Step S2, obtaining a number of actual power generation data of the current wind turbine based on a number of preset natural time periods; Step S3, defining the linear regression relationship between all actual power generation data and the actual vibration data of each component through a multivariate linear regression model; Step S4, solving the regression coefficient of the linear regression relationship by the least square method and substituting it into the multivariate linear regression model to obtain an analysis model based on the current wind turbine; Step S5, obtaining a plurality of wind speed probability distributions in the wind farm area based on a plurality of preset natural time periods; Step S6, calculating the permitted power generation data of each wind turbine based on the wind speed probability distribution of the current preset natural time period; Step S7, substituting the permitted power generation data of the current wind turbine into the analysis model for data inversion to obtain permitted vibration data based on all components of the current wind turbine; Step S8, judging whether there is outlier data in all the permitted vibration data and all the actual vibration data based on the current wind turbine, and if there is the outlier data, executing step S9; Step S9: Mark the components corresponding to all outlier data as being in abnormal operation state.

[0007] As a further improvement of the present application, step S3, defining the linear regression relationship between all actual power generation data and the actual vibration data of each component through a multivariate linear regression model, includes: Step S31, defining actual power generation data of the same preset natural time period as a dependent variable based on the current wind turbine; Step S32, defining all actual vibration data of each component in the same preset natural time period as a set of independent variables; Step S33, performing standard normalization processing on the same set of independent variables and dependent variables; Step S34, define the linear regression relationship between all dependent variables and all independent variables according to formula (1): (1); in, For the Wind turbine A dependent variable with a predetermined natural time period, is the total number of all preset natural time periods, is the intercept of the linear regression relationship, For the The linear regression coefficients of the independent variables are is the total number of independent variables in a set of independent variables, For the The first of the preset natural time periods independent variables, is the random error of the linear regression relationship.

[0008] As a further improvement of the present application, step S4, solving the regression coefficient of the linear regression relationship by the least square method and substituting it into the multivariate linear regression model, to obtain an analysis model based on the current wind turbine, including: Step S91, define the least squares method by equation (2) and solve the All regression coefficients in the linear regression relationship of the wind turbine : (2); in, for The estimated value of , is the matrix of all independent variables, ; For the matrix The transposed matrix of .

[0009] Step S92, substituting all the regression coefficients obtained into the multivariate linear regression model to obtain Analytical model of a wind turbine.

[0010] As a further improvement of the present application, step S5, obtaining a plurality of wind speed probability distributions in the wind farm area based on a plurality of preset natural time periods, includes: Step S51, obtaining a plurality of actual wind speeds in the wind farm area based on a plurality of preset natural time periods through an external wind measuring tower in the wind farm area; Step S52, define a probability distribution function (3) according to the two-parameter Weibull distribution: (3); in, is the probability distribution function, the value of which is in the interval ; is the scale parameter of the Weibull distribution; is the shape parameter of the Weibull distribution; The actual wind speed for the current preset natural time period; Step S53, defining a probability density function (4) according to the two-parameter Weibull distribution: (4); in, is the probability density function; Step S54, defining the log-likelihood function (5) of the scale parameter and the shape parameter: (5); in, is the log-likelihood function; Step S55, solving the scale parameter and the shape parameter based on the log-likelihood function; Step S56, substituting the solved scale parameter and the solved shape parameter into the probability distribution function (3) to obtain the wind speed probability distribution of the current preset natural time period.

[0011] As a further improvement of the present application, step S6, based on the wind speed probability distribution of the current preset natural time period, respectively calculates the permitted power generation data of each wind turbine, including: Step S61, define the output power relationship of the current wind turbine according to formula (6): (6); in, The wind turbine output power based on the actual wind speed of the current preset natural time period; The current wind turbine is based on the rated wind speed Rated power; is the current cut-in wind speed of the wind turbine; is the current cut-out wind speed of the wind turbine; Step S62, obtaining the running time of the current wind turbine running to the current preset natural time period; Step S63, obtaining the product of the wind turbine output power and the operating time as the permitted power generation data of the current wind turbine.

[0012] As a further improvement of the present application, step S8 is to determine whether there is outlier data in all the permitted vibration data and all the actual vibration data based on the current wind turbine. If there is outlier data, step S9 is executed, including: Step S81, integrating all permitted vibration data and all actual vibration data of the current wind turbine into a comparison data table; Step S82, sorting all the data in the comparison data table from small to large to obtain an ordered data table; Step S83, extracting outliers in the ordered data table by using the interquartile range method; Step S84: defining the abnormal value as the outlier data.

[0013] As a further improvement of the present application, step S9 marks all components corresponding to outlier data as abnormal operation states, and then includes: Step S10, obtaining a visualized operation pattern of the wind farm area; Step S20, obtaining the wind turbine position corresponding to the component marked as the abnormal operating state; Step S30, highlighting the wind turbine position in the visual operation style; Step S40, in response to an external touch operation, locally enlarging the visualized operation pattern with the wind turbine position as the center to obtain a visualized enlarged operation pattern; Step S50, loading a component visualization style for displaying all components at the wind turbine position in the visualization magnification operation style; Step S60: highlighting the components marked as being in the abnormal operating state in all component visualization styles.

[0014] In order to achieve the above objectives, this application also provides the following technical solutions: An operation analysis device for a wind turbine generator set, the operation analysis device is applied to the above-mentioned operation analysis method, and the operation analysis device comprises: An actual vibration data acquisition module, used to acquire a number of actual vibration data of each component based on a number of preset natural time periods through all vibration detection components based on the current wind turbine; An actual power generation data acquisition module is used to acquire a number of actual power generation data of the current wind turbine based on a number of preset natural time periods; A linear regression relationship definition module, used to define the linear regression relationship between all actual power generation data and actual vibration data of each component through a multivariate linear regression model; A linear regression relationship solving module, used for solving the regression coefficient of the linear regression relationship by the least square method and substituting it into the multivariate linear regression model to obtain an analysis model based on the current wind turbine; A wind speed probability distribution acquisition module, used to acquire a plurality of wind speed probability distributions in the wind farm area based on a plurality of preset natural time periods; The permitted power generation data calculation module is used to calculate the permitted power generation data of each wind turbine based on the wind speed probability distribution of the current preset natural time period; An allowed vibration data inversion module is used to substitute the allowed power generation data of the current wind turbine into the analysis model for data inversion to obtain allowed vibration data based on all components of the current wind turbine; An outlier data judging module is used to judge whether there is outlier data in all the promised vibration data and all the actual vibration data based on the current wind turbine, and if there is the outlier data, execute step S9; The abnormal operation status marking module is used to mark the components corresponding to all outlier data as being in abnormal operation status.

[0015] In order to achieve the above objectives, this application also provides the following technical solutions: An electronic device comprises a processor and a memory coupled to the processor, wherein the memory stores program instructions executable by the processor; when the processor executes the program instructions stored in the memory, an operation analysis method for a wind turbine generator set as described above is implemented.

[0016] In order to achieve the above objectives, this application also provides the following technical solutions: A storage medium stores program instructions, and when the program instructions are executed by a processor, an operation analysis method for a wind turbine generator set as described above can be implemented.

[0017] The present application is based on the current wind turbine, and obtains a number of actual vibration data of each component based on a number of preset natural time periods through all vibration detection parts; obtains a number of actual power generation data of the current wind turbine based on a number of preset natural time periods; defines the linear regression relationship between all actual power generation data and the actual vibration data of each component through a multivariate linear regression model; solves the regression coefficient of the linear regression relationship through the least squares method and substitutes it into the multivariate linear regression model to obtain an analysis model based on the current wind turbine; obtains a number of wind speed probability distributions in the wind farm area based on a number of preset natural time periods; calculates the allowed power generation data of each wind turbine based on the wind speed probability distribution of the current preset natural time period; substitutes the allowed power generation data of the current wind turbine into the analysis model for data inversion, and obtains the allowed vibration data based on all components of the current wind turbine; based on the current wind turbine, judges whether there are outlier data in all allowed vibration data and all actual vibration data. If there are outlier data, mark the components corresponding to all the outlier data as abnormal operating status. The present application utilizes the characteristic that there is a linear relationship between the vibration and power generation of the wind turbine to analyze the operating state of the wind turbine. During the operation of the wind turbine, the vibration will affect the power generation efficiency of the wind turbine, and then affect the power generation of the wind turbine. Specifically, during the operation of the wind turbine, the vibration will cause wear and fatigue of the mechanical parts, and then affect the overall performance of the wind turbine. The vibration will cause additional stress on the key parts of the wind turbine, such as the impeller and bearings, which may lead to a decrease in efficiency, thereby affecting the power generation. The wear and fatigue of the mechanical parts will produce abnormal vibration. The present application not only solves the relationship between the power generation and the vibration through linear regression, but also realizes the prediction function of power generation-vibration through linear regression. Then, the power generation that the wind turbine should have is calculated by considering the probability distribution of the wind speed of the wind farm. Then, the power generation is substituted as the dependent variable into the linear relationship for inversion to obtain the vibration that should be generated. Finally, the data obtained by forward deduction and reverse deduction are checked against each other to obtain outlier data. The outlier is abnormal. The abnormal operating state of the wind turbine is analyzed by marking the component corresponding to the abnormality. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a schematic diagram of the steps of an embodiment of a method for analyzing the operation of a wind turbine generator system according to the present application; Figure 2 This is a structural schematic diagram of an embodiment of an operation analysis device for a wind turbine generator system according to the present application; Figure 3 This is a schematic diagram of the structure of an embodiment of the electronic device of the present application; Figure 4 This is a schematic diagram of the structure of an embodiment of the storage medium of the present application. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0020] The terms "first", "second" and "third" in this application are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined as "first", "second" and "third" can explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. All directional indications (such as up, down, left, right, front, back...) in the embodiments of this application are only used to explain the relative position relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication also changes accordingly. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

[0021] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0022] like Figure 1 As shown, this embodiment provides an embodiment of an operation analysis method for a wind turbine generator set. In this embodiment, the wind turbine generator set includes a plurality of wind turbines located in the same wind farm area, and a vibration detection component is installed on each component of each wind turbine.

[0023] For example, a conventional wind turbine usually includes the following components: ① Blades: Blades are key components that capture wind energy and transfer it to the rotor axis. The design and material of blades directly affect the performance and power of wind turbines. Modern wind turbine blade materials include glass fiber, carbon fiber, etc.

[0024] ② Hub: The hub is the connection between the blade and the main shaft, responsible for transmitting the force exerted on the blade to the transmission system. The design of the hub needs to ensure sufficient strength.

[0025] ③Main shaft: The main shaft connects the blades and the generator and transmits rotational force to the generator.

[0026] ④Generator: A generator converts mechanical energy into electrical energy. Modern wind turbines usually use induction motors or asynchronous generators, with power ranging from 500 kW to 1500 kW.

[0027] ⑤Gearbox: The gearbox increases the speed of the low-speed shaft to match the speed requirement of the generator.

[0028] ⑥ Yaw system: The yaw system uses an electric motor to align the nacelle with the wind direction, ensuring that the wind turbine is always facing the direction of the wind.

[0029] ⑦ Control system: The control system monitors the status of the wind turbine and adjusts the speed, voltage and frequency of the generator as needed to ensure stable operation of the generator.

[0030] ⑧Tower: The tower supports the nacelle and rotor. Generally, the higher the better because the wind speed is greater at higher altitudes.

[0031] ⑨ Cooling system: The cooling system includes fans and oil cooling elements to keep the temperature of the generator and gearbox within an appropriate range.

[0032] ⑩‌Anemometers and wind vanes‌: These devices are used to measure wind speed and direction, helping the control system adjust the direction and power output of wind turbines.‌

[0033] For example, some of the above components can reflect some vibration characteristics at the early stage of failure. The more clear characteristics are as follows: ① The rotor mass is unbalanced. The main reasons for the unbalanced rotor mass are: uneven wear or corrosion of the impeller (mainly the blades); uneven dust accumulation or attachments (such as rust) on the blade surface; dust sticking to the hollow blades of the wings or cavities in other parts; local high temperature of the main shaft causing the shaft to bend; the impeller is not balanced after inspection; insufficient strength of the impeller causes the impeller to crack or partially deform; loose parts on the impeller or loose connectors.

[0034] The vibration characteristics of rotor mass imbalance are as follows: the vibration value is maximum in the horizontal direction and very small in the axial direction, and the vibration at the load bearing of the bearing seat is greater than that at the thrust bearing; the amplitude increases with the increase of the speed; the vibration frequency is equal to the speed frequency; the vibration stability is relatively good and is insensitive to load changes; when the hollow blades are covered with dust or individual parts are not welded firmly and displaced, the measured phase angle value is unstable, and the vibration frequency is 30% to 50% of the operating speed.

[0035] ② Vibration caused by friction between moving and static parts. The main reasons for the friction between moving and static parts are: such as collision or friction between the collector outlet and the impeller inlet, collision or friction between the impeller and the casing, and collision or friction between the main shaft and the sealing device.

[0036] The vibration characteristics of the friction between the moving and static parts are: the vibration is unstable; the vibration is self-excited and has nothing to do with the rotational speed; when the friction is severe, reverse vortex will occur.

[0037] ③ Vibration caused by abnormal rolling bearings. The main reasons for abnormal rolling bearings are: vibration caused by poor bearing assembly, vibration caused by damage to the rolling bearing surface, poor processing of the journal or shoulder, bent journal, tilted bearing installation, misalignment of the inner ring of the bearing with the axis line after assembly, causing an alternating axial force to be applied every time the bearing rotates one circle, and loosening of the fixing round nut of the rolling bearing causing local vibration; rolling bearings will be damaged by wear, rust, peeling, and fragmentation due to poor manufacturing quality, poor lubrication, entry of foreign matter, and non-standard clearance with the bearing box. The high-frequency impact vibration generated by the collision of the balls will be transmitted to the bearing seat.

[0038] The abnormal vibration characteristics of rolling bearings are as follows: the vibration value is maximum in the axial direction; the vibration frequency is equal to the rotation frequency; there is a high-frequency impact vibration signal, the vibration stability is very poor, and it is independent of the load. The vibration amplitude may be maximum in the horizontal, vertical, and axial directions. The precise diagnosis of vibration requires the help of spectrum analysis, which can accurately determine the exact location and degree of bearing damage.

[0039] ④ The vibration caused by insufficient foundation stiffness of the bearing seat. The main reasons for the insufficient foundation stiffness of the bearing seat are: poor foundation grouting, loose anchor bolts, loose gaskets, and loose machine base connections, all of which will cause severe forced resonance.

[0040] The vibration characteristics of insufficient foundation stiffness of the bearing seat are: the vibration of the bearing seat at the problematic anchor bolt is the largest, and the radial component is the largest; the vibration frequency is a combination of odd multiples of the rotational speed such as 1, 3, 5, 7, among which the 3-fold component value is the highest, which is the frequency domain characteristic of the vibration characteristic.

[0041] ⑤ Vibration caused by coupling abnormality. The main reasons for coupling abnormality are: improper installation of the coupling, eccentricity of the fan and motor shafts, failure to consider the compensation amount of axial displacement during operation when aligning the fan and motor shafts, all of which will cause vibration of the fan and motor.

[0042] The abnormal vibration characteristics of the coupling are: the vibration is irregular, changes dramatically with the load, is light when idling, and large when fully loaded, and the vibration stability is good; the greater the axis deviation, the greater the vibration; the vibration disappears when the motor runs alone; if the radial vibration is large, the two axis centerlines are parallel, and if the axial vibration is large, the two axis centerlines intersect.

[0043] It should be noted that the above vibration characteristics are only examples and do not represent all the vibration characteristics of the wind turbine. Unlisted vibration characteristics should also be taken into account. For example, when the rotor turns are short-circuited, there will be more frequency spectra with similar amplitudes, while when the stator turns are short-circuited, the amplitude changes in a pattern that increases first and then decreases.

[0044] It is worth noting that the vibration characteristics that have not yet been clarified need to be explained by the linear regression of this embodiment. The above are only examples of some characteristics, and vibrations are often accompanied by resonances of other components, resulting in a high misjudgment rate for single detection of component vibrations. However, this defect can be compensated by linear regression.

[0045] Preferably, the operation analysis method comprises the following steps: Step S1 : Based on the current wind turbine, a plurality of actual vibration data of each component based on a plurality of preset natural time periods are obtained through all vibration detection components.

[0046] Preferably, a preset natural time period can be set to a natural day, a natural hour, etc., and can be set according to actual needs.

[0047] Preferably, in actual application, when the wind turbine is generating electricity, due to the uncertainty interference of external environmental factors (unstable wind force and direction), the vibration signal of the wind turbine may be missing to varying degrees. Therefore, before detecting the vibration signal of the wind turbine, the signal collected by the vibration sensor can be subjected to conventional preprocessing.

[0048] Specifically, preprocessing can be performed by detrending terms and five-point cubic smoothing.

[0049] Among them, the detrending term is the vibration signal data collected in the vibration test. Due to the zero drift of the amplifier caused by temperature changes, the instability of the low-frequency performance outside the sensor frequency range, and the environmental interference around the sensor, it often deviates from the baseline, and even the deviation from the baseline changes with time. The whole process of the deviation from the baseline over time is called the trend term of the signal. The trend term directly affects the correctness of the signal and should be removed. The commonly used method to eliminate the trend term is the polynomial least squares method.

[0050] The detrend() function is available in MATLAB for detrending operations, but it can only remove the mean and linear trend items. Therefore, if this function is used for operation, it is assumed that the trend items contained in the sensor are linear. If the trend items are considered to be nonlinear, it is necessary to use the functions composed of polyfit() and ployval(), such as: Liu_detrend(t,y,m). In actual vibration signal data processing, polynomials of degree 1 to 3 are usually used to remove polynomial trend items from sampled data.

[0051] Among them, the five-point cubic smoothing method can be used for time domain and frequency domain signal smoothing. The main effect of this processing method on time domain data is to reduce the high-frequency random noise mixed in the vibration signal. And the effect on frequency domain data is to make the spectrum curve smooth, so as to obtain a better fitting effect in modal parameter identification. It should be noted that the five-point cubic smoothing method of frequency domain data will reduce the peak value in the spectrum curve and widen the trapezoid, which may cause the error of the identification parameter to increase.

[0052] Step S2, obtaining a plurality of actual power generation data of the current wind turbine based on a plurality of preset natural time periods.

[0053] Preferably, the actual power generation data of the wind turbine can be directly obtained through an existing monitoring system, and the acquisition method is a mature existing technology.

[0054] Step S3, defining the linear regression relationship between all actual power generation data and the actual vibration data of each component through a multivariate linear regression model.

[0055] Preferably, at this time, the dependent variable and the independent variable of the linear regression relationship are both known quantities, and the unknown quantities to be solved are the linear regression coefficients.

[0056] Step S4, solving the regression coefficient of the linear regression relationship by the least square method and substituting it into the multivariate linear regression model to obtain an analysis model based on the current wind turbine.

[0057] Preferably, the least square method is a mathematical optimization technique that seeks the best function match for data by minimizing the sum of squares of errors. The least square method can be used to easily obtain unknown data and minimize the sum of squares of errors between the obtained data and the actual data.

[0058] Step S5: obtaining a plurality of wind speed probability distributions in the wind farm area based on a plurality of preset natural time periods.

[0059] Preferably, due to the randomness of wind speed, even if all wind turbines are in the same wind farm, each wind turbine is exposed to different wind conditions, resulting in different power generation. This embodiment considers starting from the probability distribution of wind speed to ensure the rationality and accuracy of subsequent data comparison.

[0060] Step S6, respectively calculating the permitted power generation data of each wind turbine based on the wind speed probability distribution of the current preset natural time period.

[0061] Step S7, substituting the permitted power generation data of the current wind turbine into the analysis model for data inversion to obtain permitted vibration data based on all components of the current wind turbine.

[0062] Preferably, at this time, all linear regression coefficients of the linear regression relationship have been solved, that is, the linear regression coefficients are known quantities, and the allowed vibration data become unknown quantities. At this time, the allowed power generation data of the current wind turbine is substituted into the analysis model as the dependent variable to solve all independent variables, which is equivalent to a multivariate linear equation system. As long as the number of non-collinear groups of the equation system is greater than or equal to the number of variables, all unknown quantities can be solved. At this time, prepare several sets of allowed power generation data as solution support.

[0063] Step S8, judging whether there are outlier data in all the permitted vibration data and all the actual vibration data based on the current wind turbine, if there are outlier data, executing step S9.

[0064] Step S9: Mark the components corresponding to all outlier data as being in abnormal operation state.

[0065] Preferably, the methods for judging outlier data mainly include the following: ① Interquartile range method: IQR=Q3-Q1, where Q1 is the lower quartile and Q3 is the upper quartile. The upper limit is the maximum value within the non-abnormal range, and the lower limit is the minimum value within the non-abnormal range. If an observation exceeds this range, it is considered an outlier.

[0066] ②‌Raida criterion (3σ criterion): Applicable to normally distributed data. First, the mean and standard deviation of the data are calculated. Values ​​greater than the mean plus three times the standard deviation or less than the mean minus three times the standard deviation are considered outliers. This method is based on the characteristics of normal distribution, that is, most of the data (about 99.73%) will fall within the range of three times the standard deviation of the mean.

[0067] ③‌Mahalanobis distance method‌: Determine outliers in a multivariate scenario. Mahalanobis distance refers to a distance measurement in multidimensional space. If the Mahalanobis distance of an individual is greater than the critical value, it can be considered that the individual is an outlier under the test level α. The test level here is generally 0.005 or 0.001, and the calculation of the critical value is related to the test standard and the degree of freedom.

[0068] ④Median deviation method: Use the median to determine outliers. Compared with the mean, the median is not sensitive to extreme values ​​in the data, so it may be more robust in some cases. If the deviation of an observation from the median exceeds a certain threshold, it is considered an outlier.

[0069] ⑤ Analysis of variance: ANOVA is used to standardize the data, and the standard deviation is set as the boundary. Observations greater than the boundary are considered outliers. This method standardizes the data and determines outliers by comparing the relative size of the observed value and the standard deviation.

[0070] ⑥ Range analysis method: Use the range between the maximum and minimum values ​​of the data to quantitatively analyze the data. If the deviation of an observation from the maximum or minimum value exceeds a threshold calculated based on the range, it is considered an outlier.

[0071] ⑦ Statistical test method: such as Grubbs' Test and Dixon's Test. These methods are based on the principles of statistics and determine outliers by calculating statistics and comparing them with critical values.

[0072] Further, in step S3, a linear regression relationship between all actual power generation data and actual vibration data of each component is defined by a multivariate linear regression model, including: Step S31 : defining the actual power generation data of the same preset natural time period as a dependent variable based on the current wind turbine.

[0073] Step S32, defining all actual vibration data of each component in the same preset natural time period as a set of independent variables.

[0074] Step S33, performing standard normalization processing on the same set of independent variables and dependent variables.

[0075] Step S34, define the linear regression relationship between all dependent variables and all independent variables according to formula (1): (1).

[0076] in, For the Wind turbine A dependent variable with a predetermined natural time period, is the total number of all preset natural time periods, is the intercept of the linear regression relationship, For the The linear regression coefficients of the independent variables are is the total number of independent variables in a set of independent variables, For the The first of the preset natural time periods independent variables, is the random error of the linear regression relationship.

[0077] Preferably, the residual square of linear regression can be used to judge the fitting effect of the model by comparing its size. The residual sum of squares (RSS) is the sum of squares of the difference between the actual observed value and the value predicted by the regression equation, which is used to quantify the difference between the model predicted value and the actual value.

[0078] Preferably, the judgment criterion of the residual square is that the smaller the better, that is, the smaller the residual square sum is, the closer the model's predicted value is to the actual observed value, and the better the model fitting effect is; conversely, if the residual square sum is large, it indicates that there is a large deviation between the model's predicted value and the actual observed value, and the model fitting effect is not good.

[0079] Further, in step S4, the regression coefficient of the linear regression relationship is solved by the least square method and substituted into the multivariate linear regression model to obtain an analysis model based on the current wind turbine, including: Step S91, define the least squares method by equation (2) and solve the All regression coefficients in the linear regression relationship of the wind turbine : (2).

[0080] in, for The estimated value of , is the matrix of all independent variables, ; For the matrix The transposed matrix of .

[0081] Step S92, substitute all the regression coefficients obtained by solving into the multivariate linear regression model to obtain the first Analytical model of a wind turbine.

[0082] Furthermore, in step S5, a plurality of wind speed probability distributions based on a plurality of preset natural time periods in the wind farm area are obtained, including: Step S51 : obtaining a plurality of actual wind speeds in the wind farm area based on a plurality of preset natural time periods through an external wind measuring tower in the wind farm area.

[0083] Step S52, define a probability distribution function (3) according to the two-parameter Weibull distribution: (3).

[0084] in, is a probability distribution function, and the value of the probability distribution function is in the interval ; is the scale parameter of the Weibull distribution; is the shape parameter of the Weibull distribution; The actual wind speed for the current preset natural time period.

[0085] Step S53, define the probability density function (4) according to the two-parameter Weibull distribution: (4).

[0086] in, is the probability density function.

[0087] Step S54, define the log-likelihood function (5) of the scale parameter and the shape parameter: (5).

[0088] in, is the log-likelihood function.

[0089] Preferably, this embodiment provides a process for solving a log-likelihood function: set up: as well as ,but: .

[0090] Modify the above formula to get the matrix equation: .

[0091] The above matrix equation is iterated by Jacobi iteration method until the spectral radius of the matrix equation is When , it is judged to be convergent.

[0092] After convergence, the scale parameter and shape parameter of the Weibull distribution can be obtained.

[0093] Step S55, solving the scale parameter and the shape parameter based on the log-likelihood function.

[0094] Step S56, substituting the solved scale parameter and the solved shape parameter into the probability distribution function (3) to obtain the wind speed probability distribution of the current preset natural time period.

[0095] Further, step S6, based on the wind speed probability distribution of the current preset natural time period, respectively calculates the permitted power generation data of each wind turbine, including: Step S61, define the output power relationship of the current wind turbine according to formula (6): (6).

[0096] in, The wind turbine output power based on the actual wind speed of the current preset natural time period; The current wind turbine is based on the rated wind speed Rated power; is the current cut-in wind speed of the wind turbine; It is the cut-out wind speed of the current wind turbine.

[0097] Step S62, obtaining the running time of the current wind turbine running to the current preset natural time period.

[0098] Step S63, obtaining the product of the wind turbine output power and the running time as the permitted power generation data of the current wind turbine.

[0099] Further, in step S8, based on the current wind turbine, it is determined whether there is outlier data in all the promised vibration data and all the actual vibration data. If there is outlier data, step S9 is executed, including: Step S81: Integrate all permitted vibration data and all actual vibration data of the current wind turbine into a comparison data table.

[0100] Step S82, sorting all the data in the reference data table from small to large to obtain an ordered data table.

[0101] Step S83, extracting outliers in the ordered data table by using the interquartile range method.

[0102] Step S84, defining the abnormal value as outlier data.

[0103] Preferably, the interquartile range method is preferred in this embodiment to calculate outlier data. The calculation method of the interquartile range (IQR) is to subtract the third quartile (Q3) and the first quartile (Q1) of the data set, that is, IQR = Q3-Q1. The specific calculation steps are as follows: ‌Data Sorting‌: Sort the data set by numerical value.

[0104] ‌Calculate Q1‌: Find the location of 25% of the data points in the data set. The data value at this location is the first quartile (Q1). If the amount of data is an odd number, Q1 is the number before the middle two numbers; if the amount of data is an even number, Q1 is the average of the middle two numbers.

[0105] ‌Calculate Q3‌: Find the location of 75% of the data points in the data set. The data value at this location is the third quartile (Q3). Similarly, if the amount of data is odd, Q3 is the number after the two middle numbers; if the amount of data is even, Q3 is also the average of the two middle numbers, but this time it is the two numbers after.

[0106] ‌Calculate IQR‌: Use the formula IQR=Q3-Q1 to calculate the interquartile range.

[0107] The interquartile range reflects the degree of dispersion of the middle 50% of the data. The smaller the value, the more concentrated the data in the middle; the larger the value, the more dispersed the data in the middle. The interquartile range is not affected by extreme values ​​and is a robust method for measuring the degree of dispersion of data in descriptive statistics.

[0108] Further, in step S9, all components corresponding to the outlier data are marked as abnormal operation states, and then, the following steps are included: Step S10, obtaining a visualized operation pattern of the wind farm area.

[0109] Step S20, obtaining the wind turbine position corresponding to the component marked as being in abnormal operating state.

[0110] Step S30, highlighting the wind turbine position in the visual operation style.

[0111] Step S40 , in response to an external touch operation, locally enlarge the visualized operation pattern with the wind turbine position as the center to obtain a visualized enlarged operation pattern.

[0112] Step S50: loading a component visualization style for displaying all components at the wind turbine position in the visualization magnification operation style.

[0113] Step S60: highlighting the components marked as abnormal operating status in all component visualization styles.

[0114] Preferably, this embodiment further improves the user experience by visualizing abnormal components and abnormal wind turbines.

[0115] This embodiment is based on the current wind turbine, and obtains several actual vibration data of each component based on several preset natural time periods through all vibration detection parts; obtains several actual power generation data of the current wind turbine based on several preset natural time periods; defines the linear regression relationship between all actual power generation data and the actual vibration data of each component through a multivariate linear regression model; solves the regression coefficient of the linear regression relationship through the least squares method and substitutes it into the multivariate linear regression model to obtain an analysis model based on the current wind turbine; obtains several wind speed probability distributions in the wind farm area based on several preset natural time periods; calculates the allowed power generation data of each wind turbine based on the wind speed probability distribution of the current preset natural time period; substitutes the allowed power generation data of the current wind turbine into the analysis model for data inversion, and obtains the allowed vibration data based on all components of the current wind turbine; based on the current wind turbine, judges whether there are outlier data in all allowed vibration data and all actual vibration data, and if there are outlier data, marks the components corresponding to all the outlier data as abnormal operating status. This embodiment uses the characteristic that there is a linear relationship between the vibration and power generation of the wind turbine to analyze the operating state of the wind turbine. During the operation of the wind turbine, the vibration will affect the power generation efficiency of the wind turbine, and then affect the power generation of the wind turbine. Specifically, during the operation of the wind turbine, the vibration will cause wear and fatigue of mechanical parts, and then affect the overall performance of the wind turbine. The vibration will cause additional stress on key parts such as the impeller and bearings of the wind turbine, which may lead to a decrease in efficiency, thereby affecting the power generation. The wear and fatigue of mechanical parts will produce abnormal vibration. This embodiment not only solves the relationship between the power generation and vibration through linear regression, but also realizes the prediction function of power generation-vibration through linear regression. Then, the power generation of the wind turbine is calculated by considering the probability distribution of wind speed in the wind farm. Then, the power generation is substituted as the dependent variable into the linear relationship for inversion to obtain the vibration that should be generated. Finally, the data obtained by forward deduction and reverse deduction are checked against each other to obtain outlier data. The outlier is abnormal. The abnormal operating state of the wind turbine is analyzed by marking the component corresponding to the abnormality.

[0116] like Figure 2 As shown, this embodiment provides an embodiment of an operation analysis device for a wind turbine generator set. In this embodiment, the operation analysis device is applied to the operation analysis method in the above embodiment.

[0117] Specifically, the operation analysis device includes an actual vibration data acquisition module 1, an actual power generation data acquisition module 2, a linear regression relationship definition module 3, a linear regression relationship solution module 4, a wind speed probability distribution acquisition module 5, a promised power generation data calculation module 6, a promised vibration data inversion module 7, an outlier data judgment module 8, and an abnormal operation status marking module 9, which are electrically connected in sequence.

[0118] Among them, the actual vibration data acquisition module 1 is used to obtain a number of actual vibration data of each component based on a number of preset natural time periods through all vibration detection parts based on the current wind turbine; the actual power generation data acquisition module 2 is used to obtain a number of actual power generation data of the current wind turbine based on a number of preset natural time periods; the linear regression relationship definition module 3 is used to define the linear regression relationship between all actual power generation data and the actual vibration data of each component respectively through a multivariate linear regression model; the linear regression relationship solution module 4 is used to solve the regression coefficient of the linear regression relationship through the least squares method and substitute it into the multivariate linear regression model to obtain an analysis model based on the current wind turbine; the wind speed probability distribution acquisition module 5 is used to The method is used to obtain several wind speed probability distributions based on several preset natural time periods in the wind farm area; the promised power generation data calculation module 6 is used to calculate the promised power generation data of each wind turbine based on the wind speed probability distribution of the current preset natural time period; the promised vibration data inversion module 7 is used to substitute the promised power generation data of the current wind turbine into the analysis model for data inversion, and obtain the promised vibration data based on all components of the current wind turbine; the outlier data judgment module 8 is used to judge whether there are outlier data in all the promised vibration data and all the actual vibration data based on the current wind turbine; the abnormal operation status marking module 9 is used to mark the components corresponding to all the outlier data as abnormal operation status if there are outlier data.

[0119] Furthermore, the linear regression relationship definition module 3 specifically includes a first linear regression relationship definition submodule, a second linear regression relationship definition submodule, a third linear regression relationship definition submodule, and a fourth linear regression relationship definition submodule, which are electrically connected in sequence; the first linear regression relationship definition submodule is electrically connected to the actual power generation data acquisition module 2, and the fourth linear regression relationship definition submodule is electrically connected to the linear regression relationship solving module 4.

[0120] Among them, the first linear regression relationship definition submodule is used to define the actual power generation data of the same preset natural time period as a dependent variable based on the current wind turbine; the second linear regression relationship definition submodule is used to define all the actual vibration data of each component in the same preset natural time period as a group of independent variables; the third linear regression relationship definition submodule is used to perform standard normalization processing on the same group of independent variables and dependent variables; the fourth linear regression relationship definition submodule is used to define the linear regression relationship between all dependent variables and all independent variables according to formula (1): (1).

[0121] in, For the Wind turbine A dependent variable with a predetermined natural time period, is the total number of all preset natural time periods, is the intercept of the linear regression relationship, For the The linear regression coefficients of the independent variables are is the total number of independent variables in a set of independent variables, For the The first of the preset natural time periods independent variables, is the random error of the linear regression relationship.

[0122] Furthermore, the linear regression relationship solving module 4 specifically includes a first linear regression relationship solving submodule and a second linear regression relationship solving submodule which are electrically connected in sequence; the first linear regression relationship solving submodule is electrically connected to the fourth linear regression relationship definition submodule, and the second linear regression relationship solving submodule is electrically connected to the wind speed probability distribution acquisition module 5.

[0123] The first linear regression relationship solving submodule is used to define the least squares method through equation (2) and solve the first All regression coefficients in the linear regression relationship of the wind turbine : (2).

[0124] in, for The estimated value of , is the matrix of all independent variables, ; For the matrix The transposed matrix of .

[0125] The second linear regression relationship solving submodule is used to substitute all the regression coefficients obtained by solving into the multivariate linear regression model to obtain the first Analytical model of a wind turbine.

[0126] Furthermore, the wind speed probability distribution acquisition module 5 specifically includes a first wind speed probability distribution acquisition submodule, a second wind speed probability distribution acquisition submodule, a third wind speed probability distribution acquisition submodule, a fourth wind speed probability distribution acquisition submodule, a fifth wind speed probability distribution acquisition submodule, and a sixth wind speed probability distribution acquisition submodule, which are electrically connected in sequence; the first wind speed probability distribution acquisition submodule is electrically connected to the second linear regression relationship solving submodule, and the sixth wind speed probability distribution acquisition submodule is electrically connected to the promised power generation data calculation module 6.

[0127] The first wind speed probability distribution acquisition submodule is used to acquire a plurality of actual wind speeds in the wind farm area based on a plurality of preset natural time periods through an external wind measurement tower in the wind farm area.

[0128] The second wind speed probability distribution acquisition submodule is used to define the probability distribution function (3) according to the two-parameter Weibull distribution: (3).

[0129] in, is a probability distribution function, and the value of the probability distribution function is in the interval ; is the scale parameter of the Weibull distribution; is the shape parameter of the Weibull distribution; The actual wind speed for the current preset natural time period.

[0130] The third wind speed probability distribution acquisition submodule is used to define the probability density function (4) according to the two-parameter Weibull distribution: (4).

[0131] in, is the probability density function.

[0132] The fourth wind speed probability distribution acquisition submodule is used to define the log-likelihood function (5) of the scale parameter and the shape parameter: (5).

[0133] in, is the log-likelihood function.

[0134] The fifth wind speed probability distribution acquisition submodule is used to solve the scale parameter and the shape parameter based on the log-likelihood function.

[0135] The sixth wind speed probability distribution acquisition submodule is used to substitute the solved scale parameter and the solved shape parameter into the probability distribution function (3) to obtain the wind speed probability distribution of the current preset natural time period.

[0136] Furthermore, the promised power generation data calculation module 6 specifically includes a first promised power generation data calculation submodule, a second promised power generation data calculation submodule, and a third promised power generation data calculation submodule, which are electrically connected in sequence; the first promised power generation data calculation submodule is electrically connected to the sixth wind speed probability distribution acquisition submodule, and the third promised power generation data calculation submodule is electrically connected to the promised vibration data inversion module 7.

[0137] Among them, the first promised power generation data calculation submodule is used to define the output power relationship of the current wind turbine according to formula (6): (6).

[0138] in, The wind turbine output power based on the actual wind speed of the current preset natural time period; The current wind turbine is based on the rated wind speed Rated power; is the current cut-in wind speed of the wind turbine; It is the cut-out wind speed of the current wind turbine.

[0139] The second promised power generation data calculation submodule is used to obtain the operation time of the current wind turbine running to the current preset natural time period.

[0140] The third promised power generation data calculation submodule is used to obtain the product of the wind turbine output power and the operating time, which is the promised power generation data of the current wind turbine.

[0141] Furthermore, the outlier data judgment module 8 specifically includes a first outlier data judgment submodule, a second outlier data judgment submodule, a third outlier data judgment submodule, and a fourth outlier data judgment submodule, which are electrically connected in sequence; the first outlier data judgment submodule is electrically connected to the promised vibration data inversion module 7, and the fourth outlier data judgment submodule is electrically connected to the abnormal operation status marking module 9.

[0142] Among them, the first outlier data judgment submodule is used to integrate all the permitted vibration data and all the actual vibration data of the current wind turbine into a reference data table; the second outlier data judgment submodule is used to sort all the data in the reference data table from small to large to obtain an ordered data table; the third outlier data judgment submodule is used to extract outliers in the ordered data table by the interquartile range method; the fourth outlier data judgment submodule is used to define outliers as outliers.

[0143] Furthermore, the operation analysis device also includes a wind farm operation style visualization module, an abnormal wind turbine position acquisition module, an abnormal wind turbine position highlighting module, an abnormal wind turbine position touch module, a component visualization module, and an abnormal component highlighting module, which are electrically connected in sequence; the wind farm operation style visualization module is electrically connected to the abnormal operation status marking module 9.

[0144] Among them, the wind farm operation style visualization module is used to obtain the visualization operation style of the wind farm area; the abnormal wind turbine position acquisition module is used to obtain the wind turbine position corresponding to the component marked as abnormal operation status; the abnormal wind turbine position highlighting module is used to highlight the wind turbine position in the visualization operation style; the abnormal wind turbine position touch module is used to respond to external touch operations, locally enlarge the visualization operation style with the wind turbine position as the center, and obtain the visualization enlarged operation style; the component visualization module is used to load and display the component visualization style of all components at the wind turbine position of the visualization enlarged operation style; the abnormal component highlighting module is used to highlight the components marked as abnormal operation status in all component visualization styles.

[0145] It should be noted that this embodiment is a functional module embodiment based on the above method embodiment. The optimization, expansion, limitation, example, and principle description of this embodiment can be referred to the above embodiment, and this embodiment will not be repeated.

[0146] This embodiment is based on the current wind turbine, and obtains several actual vibration data of each component based on several preset natural time periods through all vibration detection parts; obtains several actual power generation data of the current wind turbine based on several preset natural time periods; defines the linear regression relationship between all actual power generation data and the actual vibration data of each component through a multivariate linear regression model; solves the regression coefficient of the linear regression relationship through the least squares method and substitutes it into the multivariate linear regression model to obtain an analysis model based on the current wind turbine; obtains several wind speed probability distributions in the wind farm area based on several preset natural time periods; calculates the allowed power generation data of each wind turbine based on the wind speed probability distribution of the current preset natural time period; substitutes the allowed power generation data of the current wind turbine into the analysis model for data inversion, and obtains the allowed vibration data based on all components of the current wind turbine; based on the current wind turbine, judges whether there are outlier data in all allowed vibration data and all actual vibration data, and if there are outlier data, marks the components corresponding to all the outlier data as abnormal operating status. This embodiment uses the characteristic that there is a linear relationship between the vibration and power generation of the wind turbine to analyze the operating state of the wind turbine. During the operation of the wind turbine, the vibration will affect the power generation efficiency of the wind turbine, and then affect the power generation of the wind turbine. Specifically, during the operation of the wind turbine, the vibration will cause wear and fatigue of mechanical parts, and then affect the overall performance of the wind turbine. The vibration will cause additional stress on key parts such as the impeller and bearings of the wind turbine, which may lead to a decrease in efficiency, thereby affecting the power generation. The wear and fatigue of mechanical parts will produce abnormal vibration. This embodiment not only solves the relationship between the power generation and vibration through linear regression, but also realizes the prediction function of power generation-vibration through linear regression. Then, the power generation of the wind turbine is calculated by considering the probability distribution of wind speed in the wind farm. Then, the power generation is substituted as the dependent variable into the linear relationship for inversion to obtain the vibration that should be generated. Finally, the data obtained by forward deduction and reverse deduction are checked against each other to obtain outlier data. The outlier is abnormal. The abnormal operating state of the wind turbine is analyzed by marking the component corresponding to the abnormality.

[0147] Figure 3 Schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 3 As shown, the electronic device 10 includes a processor 101 and a memory 102 coupled to the processor 101 .

[0148] The memory 102 stores program instructions for implementing an operation analysis method for a wind turbine generator system according to any one of the above embodiments.

[0149] The processor 101 is used to execute program instructions stored in the memory 102 to perform operation analysis on the wind turbine generator set.

[0150] The processor 101 may also be referred to as a CPU (Central Processing Unit). The processor 101 may be an integrated circuit chip having a signal processing capability. The processor 101 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0151] Further, Figure 4 This is a schematic diagram of the structure of a storage medium according to an embodiment of the present application. Figure 4 The storage medium 11 of the embodiment of the present application stores program instructions 111 that can implement all the above methods, wherein the program instructions 111 can be stored in the above storage medium in the form of a software product, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes, or terminal devices such as computers, servers, mobile phones, tablets, etc.

[0152] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0153] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. The above is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the specification and drawings of this application, or directly or indirectly used in other related technical fields, is also included in the patent protection scope of the present application.

Claims

1. A method for analyzing the operation of a wind turbine generator set, wherein the wind turbine generator set comprises a plurality of wind turbines located in the same wind farm area, and a vibration detection element is installed on each component of each wind turbine, characterized in that: The operation analysis method comprises: Step S1, based on the current wind turbine, obtaining a number of actual vibration data of each component based on a number of preset natural time periods through all vibration detection components; Step S2, obtaining a number of actual power generation data of the current wind turbine based on a number of preset natural time periods; Step S3, defining the linear regression relationship between all actual power generation data and the actual vibration data of each component through a multivariate linear regression model; Step S4, solving the regression coefficient of the linear regression relationship by the least square method and substituting it into the multivariate linear regression model to obtain an analysis model based on the current wind turbine; Step S5, obtaining a plurality of wind speed probability distributions in the wind farm area based on a plurality of preset natural time periods; Step S6, calculating the permitted power generation data of each wind turbine based on the wind speed probability distribution of the current preset natural time period; Step S7, substituting the permitted power generation data of the current wind turbine into the analysis model for data inversion to obtain permitted vibration data based on all components of the current wind turbine; Step S8, judging whether there is outlier data in all the permitted vibration data and all the actual vibration data based on the current wind turbine, and if there is the outlier data, executing step S9; Step S9: Mark the components corresponding to all outlier data as being in abnormal operation state.

2. The operation analysis method according to claim 1, characterized in that: Step S3, defining the linear regression relationship between all actual power generation data and the actual vibration data of each component through a multivariate linear regression model, including: Step S31, defining actual power generation data of the same preset natural time period as a dependent variable based on the current wind turbine; Step S32, defining all actual vibration data of each component in the same preset natural time period as a set of independent variables; Step S33, performing standard normalization processing on the same set of independent variables and dependent variables; Step S34, define the linear regression relationship between all dependent variables and all independent variables according to formula (1): （1）； in, For the Wind turbine A dependent variable with a predetermined natural time period, is the total number of all preset natural time periods, is the intercept of the linear regression relationship, For the The linear regression coefficients of the independent variables are is the total number of independent variables in a set of independent variables, For the The first of the preset natural time periods independent variables, is the random error of the linear regression relationship.

3. The operation analysis method according to claim 2, characterized in that: Step S4, solving the regression coefficient of the linear regression relationship by the least square method and substituting it into the multivariate linear regression model to obtain an analysis model based on the current wind turbine, including: Step S91, define the least squares method by equation (2) and solve the All regression coefficients in the linear regression relationship of the wind turbine : （2）； in, for The estimated value of , is the matrix of all independent variables, ; For the matrix The transposed matrix of Step S92, substituting all the regression coefficients obtained into the multivariate linear regression model to obtain Analytical model of a wind turbine.

4. The operation analysis method according to claim 1, characterized in that: Step S5, obtaining a plurality of wind speed probability distributions in the wind farm area based on a plurality of preset natural time periods, including: Step S51, obtaining a plurality of actual wind speeds in the wind farm area based on a plurality of preset natural time periods through an external wind measuring tower in the wind farm area; Step S52, define a probability distribution function (3) according to the two-parameter Weibull distribution: （3）； in, is the probability distribution function, the value of which is in the interval ; is the scale parameter of the Weibull distribution; is the shape parameter of the Weibull distribution; The actual wind speed for the current preset natural time period; Step S53, defining a probability density function (4) according to the two-parameter Weibull distribution: （4）； in, is the probability density function; Step S54, defining the log-likelihood function (5) of the scale parameter and the shape parameter: （5）； in, is the log-likelihood function; Step S55, solving the scale parameter and the shape parameter based on the log-likelihood function; Step S56, substituting the solved scale parameter and the solved shape parameter into the probability distribution function (3) to obtain the wind speed probability distribution of the current preset natural time period.

5. The operation analysis method according to claim 4, characterized in that: Step S6, respectively calculating the permitted power generation data of each wind turbine based on the wind speed probability distribution of the current preset natural time period, including: Step S61, define the output power relationship of the current wind turbine according to formula (6): （6）； in, The wind turbine output power based on the actual wind speed of the current preset natural time period; The current wind turbine is based on the rated wind speed Rated power; is the current cut-in wind speed of the wind turbine; is the current cut-out wind speed of the wind turbine; Step S62, obtaining the running time of the current wind turbine running to the current preset natural time period; Step S63, obtaining the product of the wind turbine output power and the operating time as the permitted power generation data of the current wind turbine.

6. The operation analysis method according to claim 1, characterized in that: Step S8, judging whether there is outlier data in all the promised vibration data and all the actual vibration data based on the current wind turbine, if there is the outlier data, executing step S9, including: Step S81, integrating all permitted vibration data and all actual vibration data of the current wind turbine into a comparison data table; Step S82, sorting all the data in the comparison data table from small to large to obtain an ordered data table; Step S83, extracting outliers in the ordered data table by using the interquartile range method; Step S84: defining the abnormal value as the outlier data.

7. The operation analysis method according to claim 1, characterized in that: Step S9, marking all components corresponding to outlier data as abnormal operation states, and then including: Step S10, obtaining a visualized operation pattern of the wind farm area; Step S20, obtaining the wind turbine position corresponding to the component marked as the abnormal operating state; Step S30, highlighting the wind turbine position in the visual operation style; Step S40, in response to an external touch operation, locally enlarging the visualized operation pattern with the wind turbine position as the center to obtain a visualized enlarged operation pattern; Step S50, loading a component visualization style for displaying all components at the wind turbine position in the visualization magnification operation style; Step S60: highlighting the components marked as being in the abnormal operating state in all component visualization styles.

8. An operation analysis device for a wind turbine generator set, the operation analysis device being applied to the operation analysis method according to any one of claims 1 to 7, characterized in that: The operation analysis device comprises: An actual vibration data acquisition module, used to acquire a number of actual vibration data of each component based on a number of preset natural time periods through all vibration detection components based on the current wind turbine; An actual power generation data acquisition module is used to acquire a number of actual power generation data of the current wind turbine based on a number of preset natural time periods; A linear regression relationship definition module, used to define the linear regression relationship between all actual power generation data and actual vibration data of each component through a multivariate linear regression model; A linear regression relationship solving module, used for solving the regression coefficient of the linear regression relationship by the least square method and substituting it into the multivariate linear regression model to obtain an analysis model based on the current wind turbine; A wind speed probability distribution acquisition module, used to acquire a plurality of wind speed probability distributions in the wind farm area based on a plurality of preset natural time periods; The permitted power generation data calculation module is used to calculate the permitted power generation data of each wind turbine based on the wind speed probability distribution of the current preset natural time period; An allowed vibration data inversion module is used to substitute the allowed power generation data of the current wind turbine into the analysis model for data inversion to obtain allowed vibration data based on all components of the current wind turbine; An outlier data judging module is used to judge whether there is outlier data in all the promised vibration data and all the actual vibration data based on the current wind turbine, and if there is the outlier data, execute step S9; The abnormal operation status marking module is used to mark the components corresponding to all outlier data as being in abnormal operation status.

9. An electronic device, characterized in that: It comprises a processor and a memory coupled to the processor, wherein the memory stores program instructions executable by the processor; when the processor executes the program instructions stored in the memory, an operation analysis method as described in any one of claims 1 to 7 is implemented.

10. A storage medium, characterized in that: The storage medium stores program instructions, and when the program instructions are executed by the processor, an operation analysis method according to any one of claims 1 to 7 can be implemented.

Images