CN111396265B - Fault prediction method and device of wind generating set - Google Patents
Fault prediction method and device of wind generating set Download PDFInfo
- Publication number
- CN111396265B CN111396265B CN201910004963.1A CN201910004963A CN111396265B CN 111396265 B CN111396265 B CN 111396265B CN 201910004963 A CN201910004963 A CN 201910004963A CN 111396265 B CN111396265 B CN 111396265B
- Authority
- CN
- China
- Prior art keywords
- wind
- wind speed
- generating set
- derivative variable
- wind generating
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
Images
Classifications
-
- 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
Abstract
The method for predicting the fault of the wind generating set comprises the following steps: acquiring historical wind speed data of a plurality of wind generating sets in a preset period; aiming at each wind generating set, calculating an extreme gust parameter based on historical wind speed data and design parameters of each wind generating set; establishing an extreme gust model by using the extreme gust parameter; generating a first wind speed derivative variable of each wind generating set in a preset period according to the historical wind speed data and the extreme gust model of each wind generating set; acquiring state information of each wind generating set corresponding to the first wind speed derivative variable; the state information is used for representing whether each wind generating set breaks down or not; establishing a logistic regression model according to the first wind speed derivative variable and the state information of each wind generating set; and predicting the fault probability of the wind generating set to be predicted based on the established logistic regression model and the first wind speed derivative variable of the wind generating set to be predicted.
Description
Technical Field
The invention relates to the technical field of wind power generation, in particular to a fault prediction method and device of a wind generating set.
Background
With the concern on environmental problems and the development of new energy technologies, wind power generation is more and more emphasized. Wind power generation is to convert wind energy into electric energy, the output of the electric energy is greatly influenced by wind speed, the fluctuation of the wind speed can cause the fluctuation of the output of a fan, and further influences a power grid.
At present, there are two expression methods for wind speed, the first one adopts wind speed frequency to express wind speed, and is mostly applied to wind resource assessment and wind farm design, that is, to represent the frequency of wind speed occurrence in each wind speed interval, and its calculation formula is:
wherein n is the number of wind speed sequences in a statistical observation period, and i is the number of wind speed sequences in a wind speed interval.
The second method for representing wind speed may be to simulate wind speed of a region through a weibull distribution function, which is also a common model for describing wind speed probability distribution at present, and the expression is as follows:
where x is a random variable, λ is a scale parameter, k is a shape parameter, and both λ and k are greater than 0.
The existing method for representing the wind speed has the following problems: when the wind speed frequency is adopted to represent the wind speed, the frequency of the wind speed in each wind speed interval can be represented only, the numerical value change characteristics of the wind speed in a preset period cannot be comprehensively reflected, and the mode of representing the wind speed through the wind speed frequency is over simple; although the probability distribution characteristics of the wind speed can be expressed through the Weibull distribution, the representation of the wind speed is only analyzed through the probability density distribution, so that certain limitations exist, the distribution condition and the fluctuation condition of the wind speed in a preset period cannot be measured, and further the influence of the wind speed on the fault of the wind generating set cannot be known, so that a measuring index capable of reflecting the distribution condition and the fluctuation condition of the wind speed in the preset period is urgently needed at present.
Disclosure of Invention
The invention provides a method and a device for predicting faults of a wind generating set, which can establish a prediction model related to the faults of the wind generating set and wind speed indexes according to generated indexes reflecting the distribution and fluctuation conditions of wind speed in a preset period to predict the faults of the wind generating set.
In a first aspect, the present invention provides a method for predicting a fault of a wind turbine generator system, the method comprising:
acquiring historical wind speed data of a plurality of wind generating sets in a preset period;
calculating extreme gust parameters for each wind generating set of the plurality of wind generating sets based on historical wind speed data and design parameters of each wind generating set;
establishing an extreme gust model by using the extreme gust parameter;
generating a first wind speed derivative variable of each wind generating set in the preset period according to the historical wind speed data of each wind generating set and the extreme gust model;
acquiring state information of each wind generating set corresponding to the first wind speed derivative variable; the state information is used for representing whether each wind generating set breaks down or not;
establishing a logistic regression model according to the first wind speed derivative variable and the state information of each wind generating set;
and predicting the fault probability of the wind generating set to be predicted based on the established logistic regression model and the first wind speed derivative variable of the wind generating set to be predicted.
In a second aspect, the present invention provides a method for predicting a fault of a wind turbine generator system, the method comprising:
acquiring historical wind speed data of a plurality of wind generating sets in a preset period;
aiming at each wind generating set in the plurality of wind generating sets, calculating an extreme gust parameter based on historical wind speed data and design parameters of each wind generating set;
establishing an extreme gust model by using the extreme gust parameter;
generating a first wind speed derivative variable of each wind generating set in the preset period according to the historical wind speed data of each wind generating set and the extreme gust model;
calculating a second wind speed derivative variable of each wind generating set in the preset period according to the historical wind speed data of each wind generating set;
acquiring state information of each wind generating set corresponding to the first wind speed derivative variable and the second wind speed derivative variable; the state information is used for representing whether each wind generating set breaks down or not;
establishing a logistic regression model according to the first wind speed derivative variable, the second wind speed derivative variable and the state information of each wind generating set;
and predicting the fault probability of the wind generating set to be predicted based on the established logistic regression model and the first wind speed derivative variable and the second wind speed derivative variable of the wind generating set to be predicted.
In a third aspect, the present invention provides a fault prediction apparatus for a wind turbine generator system, the apparatus comprising:
the first index generation module is used for acquiring historical wind speed data of the plurality of wind generating sets in a preset period; calculating extreme gust parameters for each wind generating set of the plurality of wind generating sets based on historical wind speed data and design parameters of each wind generating set; establishing an extreme gust model by using the extreme gust parameter; generating a first wind speed derivative variable of each wind generating set in the preset period according to the historical wind speed data of each wind generating set and the extreme gust model;
the first state information acquisition module is used for acquiring state information of each wind generating set corresponding to the first wind speed derivative variable; the state information is used for representing whether each wind generating set breaks down or not;
the first logistic regression model building module is used for building a logistic regression model according to the first wind speed derivative variable and the state information of each wind generating set;
in a fourth aspect, the present invention provides a fault prediction apparatus for a wind turbine generator system, the apparatus comprising:
the second index generation module is used for acquiring historical wind speed data of the plurality of wind generating sets in a preset period; aiming at each wind generating set in the plurality of wind generating sets, calculating an extreme gust parameter based on historical wind speed data and design parameters of each wind generating set; establishing an extreme gust model by using the extreme gust parameter; generating a first wind speed derivative variable of each wind generating set in the preset period according to the historical wind speed data of each wind generating set and the extreme gust model;
the third index generation module is used for calculating a second wind speed derivative variable of each wind generating set in the preset period according to the historical wind speed data of each wind generating set;
the second state information acquisition module is used for acquiring state information of each wind generating set corresponding to the first wind speed derivative variable and the second wind speed derivative variable; the state information is used for representing whether each wind generating set breaks down or not;
the second logistic regression model establishing module is used for establishing a logistic regression model according to the first wind speed derivative variable, the second wind speed derivative variable and the state information of each wind generating set;
and the second prediction module is used for predicting the fault probability of the wind generating set to be predicted based on the established logistic regression model and the first wind speed derivative variable and the second wind speed derivative variable of the wind generating set to be predicted.
In a fifth aspect, the present invention provides a computer readable storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the method of fault prediction for a wind park according to the first or second aspect.
In a sixth aspect, the present invention provides a computer apparatus, the apparatus comprising:
at least one processor, at least one memory, and computer program instructions stored in the memory,
the computer program instructions, when executed by the processor, implement a method of fault prediction for a wind park according to the first aspect or according to the second aspect.
According to the fault prediction method and device of the wind generating set, a first wind speed derivative variable of a plurality of wind generating sets is obtained; acquiring state information corresponding to a first wind speed derivative variable of each wind generating set in a plurality of wind generating sets; the state information is used for representing whether the plurality of wind generating sets break down or not; establishing a logistic regression model according to the first wind speed derivative variable of each wind generating set and the state information of the corresponding wind generating set; and predicting the fault probability of the wind generating set to be predicted based on the established logistic regression model and the first wind speed derivative variable of the wind generating set to be predicted. The method provided by the embodiment of the invention can predict the fault probability of the wind generating set by reflecting the related indexes of the wind speed of the distribution condition and the fluctuation condition of the wind speed in the preset period, and more effectively reflect the influence of the wind speed on the fault of the wind generating set.
Drawings
The above and other objects, features and advantages of the present invention will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which:
FIG. 1 is a flow chart of a method for generating a wind speed indicator according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for generating a wind speed indicator according to another embodiment of the present invention;
FIG. 3 is a flow chart of a method for generating a wind speed indicator according to another embodiment of the present invention;
FIG. 4 is a flow chart of a method for generating a wind speed indicator according to still another embodiment of the present invention;
FIG. 5 is a flow chart of a method for predicting a fault of a wind turbine generator system according to an embodiment of the present invention;
FIG. 6 is a flow chart of a method for predicting a fault of a wind generating set according to another embodiment of the present invention;
FIG. 7 is a schematic structural diagram of an apparatus for generating a wind speed indicator according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of a fault prediction apparatus of a wind turbine generator system according to a further embodiment of the present invention;
fig. 9 is a schematic structural diagram of a fault prediction apparatus of a wind turbine generator system according to still another embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
First embodiment
Fig. 1 is a schematic flow chart of a method for generating a wind speed indicator according to an embodiment of the present invention. As shown in fig. 1, the method for generating the wind speed index includes the following steps:
s101, obtaining historical wind speed data of the wind generating set in a preset period.
In S101, the wind turbine generator system generates a large amount of second-level instantaneous wind speed data and minute-level wind speed data during operation. Acquiring historical wind speed data generated by the wind generating set in a preset period, wherein the preset period can be a time length range of one year or two years, and the preset period can also be set according to actual requirements, such as 6 months. The historical wind speed data of the wind turbine generator system may include: at least one of second-level instantaneous wind speed data, minute-level average wind speed data and minute-level maximum wind speed data. As an example, the second-level instantaneous wind speed data may be 7-second wind speed data, and the minute-level average wind speed data may be 10-minute average wind speed data, where the 10-minute arithmetic average wind speed data may be used as the preferred minute-level average wind speed data, and the minute-level maximum wind speed data may be 10-minute maximum wind speed data, which is not limited in the embodiments of the present invention.
And S102, calculating an extreme gust parameter based on historical wind speed data and design parameters of the wind generating set.
In an exemplary embodiment, taking the instantaneous wind speed data in seconds as 7 seconds as an example, step 102 may include the following steps:
s1021, performing segmented processing on the second-level instantaneous wind speed data according to first preset time to obtain a plurality of second-level instantaneous wind speed data intervals;
the second-level instantaneous wind speed data in the historical wind speed data are processed according to a first preset time t 1 Performing a segmentation process, for example, on 7 seconds of instantaneous wind speed data according to a first predetermined time t 1 By segmentation, e.g. t 1 The time may be 10 minutes, that is, 600 seconds, starting from the first 7 seconds of instantaneous wind speed data, all 7 seconds of instantaneous wind speed data are divided into a plurality of 7 seconds of instantaneous wind speed data intervals at intervals of 10 minutes, and it should be noted that the first predetermined time t in this step is 1 The setting may be performed according to the number of the second-level instantaneous wind speed data actually obtained or the actual calculation amount, which is not limited herein.
S1022, calculating the wind speed mean value of each second-level instantaneous wind speed data interval aiming at each second-level instantaneous wind speed data interval;
after obtaining a plurality of second-level instantaneous wind speed data intervals, for example, a plurality of 7-second instantaneous wind speed data intervals, calculating a wind speed mean value in each 7-second instantaneous wind speed data interval, specifically, an arithmetic mean value of all 7-second instantaneous wind speed data in each interval, and respectively recording the obtained wind speed mean values as V hub1 、V hub2 、V hub3 …V hubn And n represents the total number of intervals.
And S1023, calculating extreme gust parameters according to the wind speed average value, the second-level instantaneous wind speed data and the design parameters of the wind generating set.
After a plurality of wind speed average values are obtained, extreme gust parameters can be calculated by utilizing the instantaneous wind speed data of each wind speed average value and second level and the design parameters of the wind generating set. Wherein the design parameters of the wind turbine generator set may comprise at least one of: impeller diameter, hub height, design turbulence value.
In an exemplary embodiment, the method of calculating the extreme gust parameter may be: according to the wind speed mean value V in the first 7-second instantaneous wind speed data interval obtained by calculation hub1 Calculating the standard deviation sigma of the wind speed of the wind generating set at the height of the hub 1 The calculation formula is as follows:
σ 1 ＝I ref *(0.75V hub1 + b) formula (3)
Wherein, b =5.6m/s, I ref The design turbulence value for the wind generating set.
Calculating to obtain the standard deviation sigma of turbulence in the first 7-second instantaneous wind speed data interval 1 And then, calculating an extreme gust parameter according to the obtained turbulence standard deviation, wherein the calculation formula is as follows:
A1＝1.35×(v e1 -v hub1 ) Formula (4)
Wherein A1 and B1 are gust coefficients, and in formula (4), v is e1 In formula (5), D is the diameter of the impeller, and Λ 1 represents a turbulence scale parameter which refers to the distance swept by a turbulent air mass in one period of rolling pulsation, and Λ is the maximum wind speed data of 7 seconds in a preset period when the height z of the hub is more than 60 meters 1 =42m; when the height z of the hub is less than or equal to 60 meters, Λ 1 =0.7 × z, the calculation formula is as follows:
wherein, the gust coefficients A1 and B1 are extreme gust parameters.
And analogizing in sequence, and respectively calculating the turbulence standard deviation in each 7-second instantaneous wind speed data interval according to the obtained n wind speed average values and the formula (3). And calculating the gust coefficient in each interval according to the n turbulence standard deviations to obtain extreme gust parameters.
And S103, establishing an extreme gust model by using the historical wind speed data and the extreme gust parameter.
The historical wind speed data and the extreme gust parameters can be utilized to establish a relation between the historical wind speed data and the extreme gust parameters obtained through calculation, and an extreme gust model is established, wherein the extreme gust model is used for representing the distribution and fluctuation conditions of the wind speed of the wind generating set in a preset period.
Taking the extreme gust parameter obtained in S1023 as an example, the method for establishing the extreme gust model may be to establish the extreme gust model according to the extreme gust parameter, the gust coefficient A1 and the gust coefficient B1 in the first 7-second instantaneous wind speed data interval obtained, and establish the extreme gust model by establishing the relationship between the gust coefficient A1 and the gust coefficient B1, which may be V gust It is shown that a specific expression of a gust model established from the relationship between the gust coefficient A1 and the gust coefficient B1 is as follows:
V gust1 = min { A1; b1 equation (7)
Equation (7) indicates that the output of the extreme gust model is the minimum of the gust coefficient A1 and the gust coefficient B1.
And S104, generating a first wind speed derivative variable of the wind generating set in a preset period according to the historical wind speed data and the extreme gust model, and taking the first wind speed derivative variable as an index for representing the wind speed.
In one exemplary embodiment, the generation method is as follows;
s1041, regarding each second-level instantaneous wind speed data interval, dividing the second-level instantaneous wind speed data interval into a plurality of second-level instantaneous wind speed data subintervals by taking second preset time as an interval;
for each at a first predetermined time t 1 For example, for the first 7-second instantaneous wind speed data interval, starting from the first 7-second instantaneous wind speed data in the interval, and recording the time as t, the first 7-second instantaneous wind speed data is recorded as v t Then for a second predetermined time t 2 The next 7 seconds of instantaneous wind speed data after interval division is v (t+t2) Obtaining a second preset time t under the second-level instantaneous wind speed data interval 2 Being a plurality of sub-intervals of space, e.g. (v) t ，v (t+t2) )、(v (t+t2) ，v (t+2*t2) )、(v (t+2*t2) ，v (t+3*t2) ) 8230, the second preset time can be set according to the second-level instantaneous wind speed data volume in each second-level instantaneous wind speed data interval, and can be divided according to the requirement of calculation accuracy. For example, when the amount of data in one section is too large, the second predetermined time may be set to be a little larger.
And S1042, calculating the difference value of the second-level instantaneous wind speed data in each subinterval at intervals of second preset time.
And calculating the difference of the 7-second instantaneous wind speed data in each subinterval according to the second-level instantaneous wind speed data division result in the S1041. The calculation method may be to calculate a difference value of instantaneous wind speed data of 7 seconds from the beginning to the end in the first subinterval, and the calculation formula is as follows:
Diff _V1 ＝V (t+t2) -V t formula (8)
And by analogy, the calculation formula of the difference value in the second subinterval is Diff _V2 ＝V (t+2*t2) -V (t+t2) Each subinterval in the second-level instantaneous wind speed data interval can be calculated according to the formula.
And S1043, judging whether the difference value meets a preset condition or not based on the extreme gust model.
According to the established extreme gust model, it is determined whether the difference in step S1042 satisfies a preset condition, where the preset condition may be, for example, whether the difference satisfies the preset condition according to a numerical relationship between the difference and the output of the extreme gust model, whether the difference and the output of the extreme gust model satisfy a set formula relationship, and the like.
S1044, recording the number of the difference values meeting the preset condition as the number of wind gust times of the wind generating set in a preset period; and taking the gust times as a first wind speed derivative variable.
Taking the example of judging whether the difference is larger than the output of the extreme gust model, V in the first 7-second instantaneous wind speed data interval is obtained gust1 For example, when Diff is used _V1 >V gust1 When the wind gust is detected, 1 gust is detected, the number of gusts detected by the wind generating set in a preset period is recorded as 1, and when the Diff is detected, the wind gust is detected _V1 >V gust1 If the wind gust is 1 time gust, recording the gust times of the wind generating set in a preset period as 2, and repeating the steps, calculating all subintervals in the first 7-second instantaneous wind speed data interval as above, and calculating the gust times in the first 7-second instantaneous wind speed data interval.
Similarly, V calculated by the method in step S102 hub2 Calculating a gust coefficient A2 and a gust coefficient B2 according to the formula (4) and the formula (5), and calculating V according to the method of the formula (7) gust2 ，
Calculating the difference values in all 7-second instantaneous wind speed data subintervals in the second interval according to the method in the step S104, and comparing the obtained difference values with V gust2 Comparing, when the difference is larger than V gust2 And when the wind is over, 1 gust of wind is experienced.
And calculating the 7-second instantaneous wind speed data intervals obtained in the step S102, and counting the data quantity of the wind gust times of all the second-level instantaneous wind speed data intervals in the predetermined period, wherein the data quantity represents the wind gust times experienced by the wind generating set in the predetermined period, and the wind gust times are used as the first wind speed derivative variable.
An embodiment of the present invention further provides a method for calculating a second wind speed derivative variable, which may be specifically referred to fig. 4, and includes one of the following steps:
step 1051, fitting Weibull distribution parameters according to minute-level average wind speed data in historical wind speed data to obtain a Weibull distribution function, and taking the Weibull distribution parameters or an average wind speed value corresponding to a probability density maximum value in the Weibull distribution function as a second wind speed derivative variable;
specifically, the acquired n 10-minute average wind speed data may be fitted to a weibull distribution function to obtain weibull distribution parameters, which are a proportional parameter λ and a shape parameter k, and the weibull distribution parameters are used as second wind speed derivative variables; the average wind speed value corresponding to the maximum probability density in the weibull distribution function may also be used as the second wind speed derivative variable.
After fitting the weibull distribution function to obtain the weibull distribution parameters, in order to verify the accuracy of the weibull distribution obtained by fitting, in an exemplary embodiment, the obtained weibull probability distribution function may be subjected to a KS test (Kolmogorov-Smirnov test, abbreviated as KS test), which is a test method based on an accumulative distribution function for testing whether a certain distribution conforms to a certain theoretical distribution or comparing whether two empirical distributions have a significant difference.
When the obtained weibull probability distribution function is valid, an average wind speed value corresponding to a maximum value of the probability density in the weibull distribution function may be used as the second wind speed derivative variable.
Specifically, the difference between the minute-scale average wind speed data and the average wind speed value corresponding to the maximum probability density value in the weibull distribution function may be calculated, and the difference may be used as the second wind speed derivative variable.
Step 1052, calculating a minute-level average wind speed value of the wind generating set in a preset period according to minute-level average wind speed data in the historical wind speed data, and taking the minute-level average wind speed value in the preset period as a second wind speed derivative variable.
In one exemplary embodiment, a 10-minute average wind speed value of the wind generating set in a preset period is calculated according to 10-minute average wind speed data in historical wind speed data; and taking the average wind speed value of 10 minutes in the preset period as a second wind speed derivative variable. The specific method can be as follows: calculating the average wind speed in a preset period according to the average wind speed in 10 minutes, wherein the calculation formula is as follows:
wherein v is i Represents the ith 10-minute average wind speed, and n represents the number of 10-minute average wind speeds acquired within 1 year in a predetermined period of 1 year as an example; the average wind speed of the wind generating set in 1 year can be calculated through the formula (9).
And 1053, calculating a minute-level maximum wind speed value of the wind generating set in a preset period according to minute-level maximum wind speed data in the historical wind speed data, and taking the minute-level maximum wind speed value in the preset period as a second wind speed derivative variable.
In one exemplary embodiment, the maximum wind speed within the predetermined period may be calculated from the 10 minute maximum wind speed data in the historical wind speed data, with the 10 minute maximum wind speed value as the first wind speed derivative variable. The specific calculation formula is as follows:
v _ max = max (V _ max1, V _ max2, V _ max3 \ 8230; \8230; V _ maxn) (equation 10)
Wherein V _ max1 represents the 1 st 10-minute maximum wind speed, V _ max2 represents the 2 nd 10-minute maximum wind speed, and so on, V _ maxn represents the nth 10-minute maximum wind speed, and V _ max is the calculated maximum wind speed in the predetermined period.
Step 1054, partitioning the minute-level average wind speed data in the historical wind speed data, and calculating the wind speed number in each partition; and calculating the unit wind quantity of the wind generating set in a preset period according to the wind speed quantity in each partition, and taking the unit wind quantity as a second wind speed derivative variable.
In one exemplary embodiment, 10-minute average wind speed data in the historical wind speed data can be partitioned, and the wind speed number in each partition is calculated; calculating the unit wind quantity of the wind generating set in a preset period according to the wind speed quantity in each subarea; and taking the unit wind quantity as a second wind speed derivative variable. The specific partitioning method may be to select the maximum value and the minimum value in the 10-minute average wind speed in the historical wind speed data in the predetermined period, divide the maximum value and the minimum value by taking 1 as a unit, calculate the wind speed number in each interval, and perform multiplication calculation on the interval median value in each interval and the wind speed number in the interval to obtain the experienced unit wind number of the wind turbine generator set in the interval.
The specific method for calculating the unit wind quantity of the wind generating set in the preset period according to the wind speed quantity in each subarea can be as follows: calculating a median of the 10 minute average wind speed data within each partition; calculating the unit wind quantity in each subarea according to the median value of the interval and the wind speed quantity in each subarea; and calculating the unit wind quantity of the wind generating set in a preset period according to the unit wind quantity in each partition.
Table 1 shows the number of wind units of a wind turbine generator set in each interval.
TABLE 1
| Interval of wind speeds (m/s) | Median value of interval | Number of wind speeds | Number of unit wind in interval |
| (0.5～1.5] | 1 | 2076 | 2076 |
| (1.5～2.5] | 2 | 3648 | 7296 |
| (2.5～3.5] | 3 | 4104 | 12312 |
| (3.5～4.5] | 4 | 5268 | 21072 |
| (4.5～5.5] | 5 | 6036 | 30180 |
| (5.5～6.5] | 6 | 6804 | 40824 |
| (6.5～7.5] | 7 | 6600 | 46200 |
| (7.5～8.5] | 8 | 5652 | 45216 |
| (8.5～9.5] | 9 | 4572 | 41148 |
| (9.5～10.5] | 10 | 3360 | 33600 |
| (10.5～11.5] | 11 | 2892 | 31812 |
| (11.5～12.5] | 12 | 1548 | 18576 |
| (12.5～13.5] | 13 | 600 | 7800 |
| (13.5～14.5] | 14 | 252 | 3528 |
| (14.5～15.5] | 15 | 48 | 720 |
| (15.5～16.5] | 16 | 24 | 384 |
As can be seen from table 1, when the 10-minute average wind speed in the predetermined period is 0.7m/s at the minimum and 15.7m/s at the maximum, the unit wind number in the interval = the median wind speed number in the interval, and the unit wind number in the predetermined period is obtained by adding the sum of the unit wind numbers in all the intervals, for example, when the predetermined period is 1 year, the unit wind number in 1 year is obtained by adding the unit wind numbers in all the intervals.
After the calculation as described above, the weibull distribution scale parameter λ and the shape parameter k, the maximum wind speed data in a predetermined period, the average wind speed value corresponding to the maximum probability density in the weibull probability distribution function, and the difference Δ between the 10min average wind speed and the average wind speed value corresponding to the maximum probability density in the weibull probability distribution function may be calculated wspd The unit wind quantity and the gust frequency in the preset period are used as multi-index representation results of wind speed, and the wind speed distribution condition and the fluctuation condition of the wind generating set in the preset period can be represented.
Taking table 2 as an example, table 2 shows the result of performing multi-index calculation of wind speed on 10 wind turbine generators:
TABLE 2
In table 2, the data results of the multiple indexes representing the wind speed calculated according to the historical wind speed data for 10 wind turbine generators can comprehensively reflect the digital characteristics of the wind speed from 8 dimensions, and support can be provided for establishing a wind turbine generator system fault early warning model based on the multiple indexes of the wind speed in the later period.
In a second aspect, an embodiment of the present invention provides a method for predicting a fault of a wind turbine generator system, including the following steps, as shown in fig. 5:
step 201, obtaining a first wind speed derivative variable generated by the above method for a plurality of wind generating sets.
And acquiring a first wind speed derivative variable of the plurality of wind generating sets, wherein the generation method of the first wind speed derivative variable is the generation method of the wind speed index.
Step 202, acquiring state information corresponding to a first wind speed derivative variable of each wind generating set in a plurality of wind generating sets; the state information is used for representing whether the plurality of wind generating sets break down or not.
After the first wind speed derivative variable of each wind generating set is obtained, state information of each wind generating set is correspondingly obtained, the first wind speed derivative variable of each wind generating set corresponds to the state information one by one, wherein the state information of each wind generating set is information of whether the wind generating set fails, and the failure information can include information of whether each component of the wind generating set fails or is damaged.
And step 203, establishing a logistic regression model according to the first wind speed derivative variable of each wind generating set and the state information of the corresponding wind generating set.
And establishing a logistic regression model by taking the information of whether the fault or the damage occurs as a dependent variable and the first wind speed derivative variable as an independent variable according to the first wind speed derivative variable and the corresponding state information of each wind generating set.
In an exemplary embodiment, the method for establishing the logistic regression model may be to perform fitting calculation on the first wind speed derivative variables of all the wind generating sets and the state information of the corresponding wind generating sets to obtain modeling parameters of the logistic regression model; the modeling parameters may include a slope parameter and an intercept parameter, the slope parameter is a coefficient of the first wind speed derivative variable, and the intercept parameter is a constant in a logistic regression model obtained after fitting calculation.
And establishing a corresponding relation between the first wind speed derivative variable and the fault probability of the wind generating set based on the modeling parameters to generate a logistic regression model.
And establishing a corresponding relation between the first wind speed derivative variable and the fault probability of the wind turbine generator according to the obtained modeling parameters, namely the slope parameter and the intercept parameter, and generating a logistic regression model based on the corresponding relation.
And step 204, predicting the fault probability of the wind generating set to be predicted based on the established logistic regression model and the first wind speed derivative variable of the wind generating set to be predicted.
Specifically, the method for predicting the fault probability of the wind turbine generator system to be predicted may be: the method comprises the steps of obtaining a first wind speed derivative variable of a wind generating set to be predicted, inputting the first wind speed derivative variable of the wind generating set to be predicted into a logistic regression model, and obtaining the fault probability of the wind generating set to be predicted according to the established corresponding relation between the first wind speed derivative variable and the fault probability of the wind generating set.
In an exemplary embodiment, the above failure prediction analysis may be performed on the wind speed multi-index calculation results of the 10 wind turbine generators, and 8 indexes and status information of the 10 wind turbine generators are as follows, and the results are shown in table 3:
TABLE 3
The multiple wind speed indexes obtained by the embodiment can provide support for establishing a digital model for the fault early warning of the wind generating set.
Then, aiming at 8 indexes given in the table 3, a logistic regression model is established by taking the gust times as an example, and the gust times and the state information of whether the gust times are damaged are taken as variables, wherein 'whether the gust times are damaged' in the table 3 is 0, which indicates that the wind generating set is not damaged; the "damage or not" is 1, which indicates that the wind turbine generator system is damaged. The number of gusts and whether the gusts are damaged are used as variables, a logistic regression model is established, and the obtained parameters are shown in table 4:
TABLE 4
It can be seen from table 4 that the significance test value P of the logistic regression model established according to the number of gust is much less than 0.05, which proves that the model has validity, and the regression model parameters are as follows: the slope parameter is 0.00038, the intercept parameter is-3.34800, and the logistic regression equation is:
ln (p/(1-p)) =0.00038 × gust number-3.348 equation (11)
According to the original logistic regression equation:
ln (p/(1-p)) = a x + b formula (12)
In the present embodiment, p in the formula (12) is a predicted probability value of a certain event, which represents a failure probability of the wind turbine generator system, a is a slope parameter, b is an intercept parameter, and x is a variable, which represents a first wind speed derivative variable in the embodiment of the present invention, and may be any one of the indexes in table 3.
In the logistic regression model, when the value of x is increased, for example, when x is increased by 1, the result of comparing the logistic regression equation after the variable is increased by 1 with the original logistic regression equation is:
and (3) performing deformation calculation on the formula (13) to obtain a formula (14):
according to the formula (14), when the number of wind gust times is increased by 100, the component failure probability of the wind generating set is changed to be 1.038 times, and the failure probability is increased by 3.8%, so that the influence of the number of wind gust times on the component failure of the wind generating set can be reflected.
Similarly, in this embodiment, a logistic regression model may be established for other indexes, for example, a logistic regression model is established for the unit wind quantity, and the parameter table obtained by establishing the logistic regression model with the unit wind quantity and the damaged state information as variables is shown in the following table 5:
TABLE 5
The p value of the model test is far less than 0.05, the validity of the model is proved, and the logistic regression equation is as follows:
ln (p/(1-p)) =0.000008 × unit wind quantity-5.141669 equation (15)
According to the formula (15), the formula (14) and the formula (13), the damage probability of most parts becomes 1.0080 times of the original damage probability every time the number of the unit wind is increased by 1000, namely the damage probability is increased by 0.80%, so that the influence of the number of the unit wind on the part failure of the wind generating set can be reflected.
In a third aspect, an embodiment of the present invention further provides a method for predicting a fault of a wind turbine generator system, including the following steps, as shown in fig. 6:
step 301, acquiring a first wind speed derivative variable generated by the method of the plurality of wind generating sets;
and acquiring a first wind speed derivative variable of the plurality of wind generating sets, wherein the generation method of the first wind speed derivative variable is the generation method of the wind speed index.
Step 302, acquiring a second wind speed derivative variable generated by the method of the plurality of wind generating sets;
and acquiring a second wind speed derivative variable of the plurality of wind generating sets, wherein the generation method of the second wind speed derivative variable is the generation method of the wind speed index.
303, acquiring state information corresponding to a first wind speed derivative variable and a second wind speed derivative variable of each wind generating set in the plurality of wind generating sets; the state information is used for representing whether the plurality of wind generating sets break down or not;
after the first wind speed derivative variable and the second wind speed derivative variable of each wind generating set are obtained, state information of each wind generating set is correspondingly obtained, the first wind speed derivative variable, the second wind speed derivative variable and the state information of each wind generating set correspond to each other one by one, wherein the state information of each wind generating set is information of whether the wind generating set fails or not, and the failure information can include information of whether each component of the wind generating set fails or is damaged.
Step 304, establishing a logistic regression model according to the first wind speed derivative variable, the second wind speed derivative variable and the state information of the wind generating set;
in this step, the method for establishing the logistic regression model may be: fitting and calculating the first wind speed derivative variable, the second wind speed derivative variable and the state information of each wind generating set to obtain modeling parameters of a logistic regression model; the modeling parameters may include a slope parameter and an intercept parameter, the slope parameter is a coefficient before the first wind speed derivative variable and the second wind speed derivative variable, and the intercept parameter is a constant in the logistic regression model obtained after fitting calculation.
And establishing a corresponding relation by taking the first wind speed derivative variable and the second wind speed derivative variable as independent variables and taking the fault probability of the wind generating set as a dependent variable based on the modeling parameters to generate a logistic regression model.
And 305, predicting the fault probability of the wind generating set to be predicted based on the established logistic regression model and the first wind speed derivative variable and the second wind speed derivative variable of the wind generating set to be predicted.
Specifically, the method for predicting the fault probability of the wind turbine generator system to be predicted may be: acquiring a first wind speed derivative variable and a second wind speed derivative variable of a wind generating set to be predicted;
and inputting the first wind speed derivative variable and the second wind speed derivative variable of the wind generating set to be predicted into a logistic regression model, and obtaining the fault probability of the wind generating set to be predicted according to the established corresponding relation which takes the first wind speed derivative variable and the second wind speed derivative variable as independent variables and takes the fault probability of the wind generating set as dependent variables.
In an exemplary embodiment, taking λ and k indexes in table 3 as examples, a logistic regression model is established, and a parameter table obtained by establishing the logistic regression model is shown in table 6:
TABLE 6
The p value of the model is far less than 0.05 through model test, the validity of the model is proved, and the logistic regression equation is as follows:
ln (p/(1-p)) =0.00566 λ +0.01011 k-2.03094 equation (16)
As can be seen from the formulas (16), (14) and (13), for every 1 λ increase, the large-part damage probability becomes 1.0057 times, i.e., increases by 0.57%, and for every 1 k increase, the large-part damage probability becomes 1.0102 times, i.e., increases by 1.02%.
The influence of λ and k on the probability of failure of the wind turbine can thus be reflected.
In an exemplary embodiment, a logistic regression model may be established by taking two indexes of λ and wind gust times in table 3 as an example, and a parameter table obtained by establishing the logistic regression model is shown in table 7 below:
TABLE 7
The p value of the model is far less than 0.05 through model test, the validity of the model is proved, and the logistic regression equation is as follows:
ln (p/(1-p)) =0.00566 λ +0.00017 λ gust times-2.03094 equation (17)
According to the formula (17), the formula (14) and the formula (13), the fault probability of the wind generating set is changed to 1.0057 times of the original fault probability when the lambda is increased by 1, namely the fault probability is increased by 0.57%; the failure probability is changed to 1.0274 times, namely 2.74% for every 100 times of increase of the gust frequency.
Therefore, the influence of lambda and the wind gust frequency on the fault probability of the wind generating set can be reflected.
An embodiment of the present invention further provides a device 400 for generating a wind speed indicator, where a schematic structural diagram is shown in fig. 7, and the device includes:
the acquiring module 401 is configured to acquire historical wind speed data of the wind turbine generator system in a predetermined period;
a calculating module 402, configured to calculate an extreme gust parameter based on historical wind speed data and design parameters of the wind turbine generator set;
a model establishing module 403, configured to establish an extreme gust model by using the extreme gust parameter;
and a first index generating module 404, configured to generate a first wind speed derivative variable of the wind turbine generator set in a predetermined period according to the historical wind speed data and the extreme gust model, and use the first wind speed derivative variable as an index representing a wind speed.
In one exemplary embodiment, the calculation module 402 includes:
the segmentation submodule is used for carrying out segmentation processing on the second-level instantaneous wind speed data according to first preset time to obtain a plurality of second-level instantaneous wind speed data intervals;
the first calculation submodule is used for calculating the wind speed mean value of each second-level instantaneous wind speed data interval aiming at each second-level instantaneous wind speed data interval;
the second calculation submodule is used for calculating extreme gust parameters according to the wind speed mean value, the second-level instantaneous wind speed data and the design parameters of the wind generating set;
wherein the design parameters of the wind generating set may comprise at least one of: impeller diameter, hub height, design turbulence value.
In an exemplary embodiment, the second calculation submodule is specifically configured to calculate a turbulence standard deviation of the wind turbine generator set within each second-level instantaneous wind speed data interval according to each wind speed mean value and the design turbulence value in the design parameter;
and calculating an extreme gust parameter according to the turbulence standard deviation, the impeller diameter and the hub height.
In an exemplary embodiment, the first index generation module 404 is specifically configured to divide the second-level instantaneous wind speed data interval into a plurality of second-level instantaneous wind speed data subintervals at intervals of a second predetermined time for each second-level instantaneous wind speed data interval;
calculating the difference value of the second-level instantaneous wind speed data in each subinterval at intervals of second preset time; judging whether the difference value meets a preset condition or not based on the extreme gust model;
recording the number of the difference values meeting the preset condition as the number of gust times of the wind generating set in a preset period;
and taking the gust times as a first wind speed derivative variable.
In one exemplary embodiment, the apparatus further comprises:
a second index generation module 405, configured to calculate a second wind speed derivative variable according to the historical wind speed data; the second wind speed derivative variable is used as an index representing the wind speed.
In an exemplary embodiment, the second index generation module 405 is configured to perform at least one of the following steps:
fitting a Weibull distribution parameter according to the minute-level average wind speed data in the historical wind speed data to obtain a Weibull distribution function, and taking the Weibull distribution parameter or an average wind speed value corresponding to the probability density maximum value in the Weibull distribution function as a second wind speed derivative variable;
calculating a minute-level average wind speed value of the wind generating set in a preset period according to minute-level average wind speed data in the historical wind speed data, and taking the minute-level average wind speed value in the preset period as a second wind speed derivative variable;
calculating a minute-level maximum wind speed value of the wind generating set in a preset period according to minute-level maximum wind speed data in the historical wind speed data, and taking the minute-level maximum wind speed value in the preset period as a second wind speed derivative variable;
partitioning the minute-level average wind speed data in the historical wind speed data, and calculating the wind speed number in each partition; and calculating the unit wind quantity of the wind generating set in a preset period according to the wind speed quantity in each subarea, and taking the unit wind quantity as a second wind speed derivative variable.
In an exemplary embodiment, the second index generation module is specifically configured to calculate a median value of the minute-scale average wind speed data within said each partition;
calculating the unit wind quantity in each subarea according to the median value and the wind speed quantity in each subarea;
and calculating the unit wind quantity of the wind generating set in a preset period according to the unit wind quantity in each subarea.
The invention also provides a failure prediction device 500 of a wind generating set, the structural schematic diagram of the device is shown in fig. 8, and the device comprises:
the first index generation module 501 is configured to obtain historical wind speed data of a plurality of wind turbine generators in a predetermined period; calculating extreme gust parameters for each wind generating set of the plurality of wind generating sets based on historical wind speed data and design parameters of each wind generating set; establishing an extreme gust model by using the extreme gust parameter; generating a first wind speed derivative variable of each wind generating set in a preset period according to the historical wind speed data and the extreme gust model of each wind generating set;
a first state information obtaining module 502, configured to obtain, for each wind turbine generator set, state information corresponding to a first wind speed derivative variable; the state information is used for judging whether each wind generating set fails or not;
a first logistic regression model establishing module 503, configured to establish a logistic regression model according to the first wind speed derivative variable and the state information of each wind turbine generator set;
the first prediction module 504 is configured to predict a failure probability of the wind turbine generator system to be predicted based on the established logistic regression model and the first wind speed derivative variable of the wind turbine generator system to be predicted.
In one exemplary embodiment, the first logistic regression model establishing module 503 includes:
the first fitting calculation submodule is used for performing fitting calculation on the first wind speed derivative variable and the state information of each wind generating set to obtain a modeling parameter of the logistic regression model;
and the first model generation submodule is used for establishing a corresponding relation between the first wind speed derivative variable and the fault probability of the wind generating set based on the modeling parameters and generating a logistic regression model.
In an exemplary embodiment, the first prediction module 504 is specifically configured to obtain a first wind speed derivative variable of the wind turbine generator set to be predicted; and inputting the first wind speed derivative variable of the wind generating set to be predicted into a logistic regression model, and obtaining the fault probability of the wind generating set to be predicted according to the established corresponding relation between the first wind speed derivative variable and the fault probability of the wind generating set.
The invention also provides a failure prediction device 600 of a wind generating set, the structural schematic diagram of the device is shown in fig. 9, and the device comprises:
the second index generation module 601 is configured to obtain historical wind speed data of the plurality of wind turbine generators in a predetermined period; aiming at each wind generating set in the plurality of wind generating sets, calculating an extreme gust parameter based on historical wind speed data and design parameters of each wind generating set; establishing an extreme gust model by using the extreme gust parameter; generating a first wind speed derivative variable of each wind generating set in a preset period according to the historical wind speed data and the extreme gust model of each wind generating set;
a third index generating module 602, configured to calculate a second wind speed derivative variable of each wind generating set in a predetermined period according to the historical wind speed data of each wind generating set;
a second state information obtaining module 603, configured to obtain state information of each wind turbine generator system corresponding to the first wind speed derivative variable and the second wind speed derivative variable; the state information is used for representing whether the plurality of wind generating sets break down or not;
a second logistic regression model establishing module 604, configured to establish a logistic regression model according to the first wind speed derivative variable, the second wind speed derivative variable, and the state information of each wind turbine generator set;
and a second prediction module 605, configured to predict a failure probability of the wind generating set to be predicted based on the established logistic regression model and the first wind speed derivative variable and the second wind speed derivative variable of the wind generating set to be predicted.
In an exemplary embodiment, the second logistic regression model building module 604 includes:
the second fitting calculation submodule is used for performing fitting calculation on the first wind speed derivative variable, the second wind speed derivative variable and the state information of each wind generating set to obtain a modeling parameter of the logistic regression model;
and the second model generation submodule is used for establishing a corresponding relation which takes the first wind speed derivative variable and the second wind speed derivative variable as independent variables and takes the fault probability of the wind generating set as a dependent variable based on the modeling parameters to generate a logistic regression model.
In an exemplary embodiment, the second prediction module 605 is specifically configured to obtain a first wind speed derivative variable and a second wind speed derivative variable of the wind turbine generator set to be predicted;
and inputting the first wind speed derivative variable and the second wind speed derivative variable of the wind generating set to be predicted into a logistic regression model, and obtaining the fault probability of the wind generating set to be predicted according to the established corresponding relation which takes the first wind speed derivative variable and the second wind speed derivative variable as independent variables and takes the fault probability of the wind generating set as dependent variables.
Embodiments of the present invention also provide a computer-readable storage medium, on which computer program instructions are stored, which, when executed by a processor, implement the wind speed index generation method as described above.
Embodiments of the present invention also provide a computer readable storage medium, on which computer program instructions are stored, which when executed by a processor implement the method for predicting a fault of a wind turbine generator system according to the second aspect.
Embodiments of the present invention also provide a computer readable storage medium, on which computer program instructions are stored, which when executed by a processor implement the method for predicting a fault of a wind turbine generator system according to the third aspect.
An embodiment of the present invention further provides a computer apparatus, where the apparatus includes:
at least one processor, at least one memory, and computer program instructions stored in the memory,
the computer program instructions, when executed by a processor, implement the wind speed indicator generation method as described above.
An embodiment of the present invention further provides a computer apparatus, including:
at least one processor, at least one memory, and computer program instructions stored in the memory,
the computer program instructions, when executed by the processor, implement a method of fault prediction for a wind park according to the second aspect.
An embodiment of the present invention further provides a computer apparatus, including:
at least one processor, at least one memory, and computer program instructions stored in the memory,
the computer program instructions, when executed by the processor, implement a method of fault prediction for a wind park as described in the third aspect.
It should be noted that the apparatuses in the foregoing embodiments can be used as the execution main body in the methods in the foregoing embodiments, and can implement corresponding processes in the methods to achieve the same technical effects, and for brevity, the contents of this aspect are not described again.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment may be implemented by software plus a necessary general hardware platform, and may also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods of the various embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (14)
1. A method for predicting a fault of a wind generating set, the method comprising:
acquiring historical wind speed data of a plurality of wind generating sets in a preset period;
aiming at each wind generating set in the plurality of wind generating sets, calculating an extreme gust parameter based on historical wind speed data and design parameters of each wind generating set;
establishing an extreme gust model by using the extreme gust parameter;
generating a first wind speed derivative variable of each wind generating set in the preset period according to the historical wind speed data of each wind generating set and the extreme gust model, wherein the first wind speed derivative variable is an index reflecting the distribution and fluctuation condition of the wind speed in the preset period;
acquiring state information of each wind generating set corresponding to the first wind speed derivative variable; the state information is used for representing whether each wind generating set breaks down or not;
establishing a logistic regression model according to the first wind speed derivative variable and the state information of each wind generating set;
and predicting the fault probability of the wind generating set to be predicted based on the established logistic regression model and the first wind speed derivative variable of the wind generating set to be predicted.
2. The method of claim 1, wherein the step of building a logistic regression model based on the first wind speed derived variables and state information for each wind turbine generator set comprises:
fitting calculation is carried out on the first wind speed derivative variable and the state information of each wind generating set, and modeling parameters of the logistic regression model are obtained;
and establishing a corresponding relation between the first wind speed derivative variable and the fault probability of the wind generating set based on the modeling parameters, and generating the logistic regression model.
3. The method of claim 2, wherein the step of predicting the failure probability of the wind park to be predicted based on the established logistic regression model and the first wind speed derivative variable of the wind park to be predicted comprises:
acquiring a first wind speed derivative variable of a wind generating set to be predicted;
and inputting the first wind speed derivative variable of the wind generating set to be predicted into the logistic regression model, and obtaining the fault probability of the wind generating set to be predicted according to the established corresponding relation between the first wind speed derivative variable and the fault probability of the wind generating set.
4. A method for predicting a fault of a wind generating set, the method comprising:
acquiring historical wind speed data of a plurality of wind generating sets in a preset period;
calculating extreme gust parameters for each wind generating set of the plurality of wind generating sets based on historical wind speed data and design parameters of each wind generating set;
establishing an extreme gust model by using the extreme gust parameter;
generating a first wind speed derivative variable of each wind generating set in the preset period according to the historical wind speed data of each wind generating set and the extreme gust model, wherein the first wind speed derivative variable is an index reflecting the distribution and fluctuation condition of the wind speed in the preset period;
calculating a second wind speed derivative variable of each wind generating set in the preset period according to the historical wind speed data of each wind generating set;
acquiring state information of each wind generating set corresponding to the first wind speed derivative variable and the second wind speed derivative variable; the state information is used for representing whether each wind generating set breaks down or not;
establishing a logistic regression model according to the first wind speed derivative variable, the second wind speed derivative variable and the state information of each wind generating set;
and predicting the fault probability of the wind generating set to be predicted based on the established logistic regression model and the first wind speed derivative variable and the second wind speed derivative variable of the wind generating set to be predicted.
5. The method of claim 4, wherein the step of building a logistic regression model based on the first wind speed derived variable, the second wind speed derived variable, and the state information for each wind turbine generator set comprises:
fitting and calculating the first wind speed derivative variable, the second wind speed derivative variable and the state information of each wind generating set to obtain modeling parameters of the logistic regression model;
and establishing a corresponding relation by taking the first wind speed derivative variable and the second wind speed derivative variable as independent variables and taking the fault probability of the wind generating set as dependent variables based on the modeling parameters to generate the logistic regression model.
6. The method of claim 5, wherein the step of predicting the failure probability of the wind park to be predicted based on the established logistic regression model and the first and second wind speed derived variables of the wind park to be predicted comprises:
acquiring a first wind speed derivative variable and a second wind speed derivative variable of a wind generating set to be predicted;
and inputting the first wind speed derivative variable and the second wind speed derivative variable of the wind generating set to be predicted into the logistic regression model, and obtaining the fault probability of the wind generating set to be predicted according to the established corresponding relation which takes the first wind speed derivative variable and the second wind speed derivative variable as independent variables and takes the fault probability of the wind generating set as dependent variables.
7. A fault prediction device for a wind turbine generator system, the device comprising:
the first index generation module is used for acquiring historical wind speed data of the plurality of wind generating sets in a preset period; aiming at each wind generating set in the plurality of wind generating sets, calculating an extreme gust parameter based on historical wind speed data and design parameters of each wind generating set; establishing an extreme gust model by using the extreme gust parameter; generating a first wind speed derivative variable of each wind generating set in the preset period according to the historical wind speed data of each wind generating set and the extreme gust model, wherein the first wind speed derivative variable is an index reflecting the distribution and fluctuation condition of the wind speed in the preset period;
the first state information acquisition module is used for acquiring state information of each wind generating set corresponding to the first wind speed derivative variable; the state information is used for representing whether each wind generating set breaks down or not;
the first logistic regression model building module is used for building a logistic regression model according to the first wind speed derivative variable and the state information of each wind generating set;
the first prediction module is used for predicting the fault probability of the wind generating set to be predicted based on the established logistic regression model and the first wind speed derivative variable of the wind generating set to be predicted.
8. The apparatus according to claim 7, wherein the first logistic regression model establishing module is specifically configured to perform fitting calculation on the first wind speed derivative variable and the state information of each wind turbine generator set to obtain modeling parameters of the logistic regression model;
and establishing a corresponding relation between the first wind speed derivative variable and the fault probability of the wind generating set based on the modeling parameters, and generating the logistic regression model.
9. The device of claim 8, wherein the first prediction module is specifically configured to obtain a first wind speed derivative variable of the wind turbine generator set to be predicted; and inputting the first wind speed derivative variable of the wind generating set to be predicted into the logistic regression model, and obtaining the fault probability of the wind generating set to be predicted according to the established corresponding relation between the first wind speed derivative variable and the fault probability of the wind generating set.
10. A fault prediction device for a wind turbine generator system, the device comprising:
the second index generation module is used for acquiring historical wind speed data of the plurality of wind generating sets in a preset period; aiming at each wind generating set in the plurality of wind generating sets, calculating an extreme gust parameter based on historical wind speed data and design parameters of each wind generating set; establishing an extreme gust model by using the extreme gust parameter; generating a first wind speed derivative variable of each wind generating set in the preset period according to the historical wind speed data of each wind generating set and the extreme gust model, wherein the first wind speed derivative variable is an index reflecting the distribution and fluctuation condition of the wind speed in the preset period;
the third index generation module is used for calculating a second wind speed derivative variable of each wind generating set in the preset period according to the historical wind speed data of each wind generating set;
the second state information acquisition module is used for acquiring state information of each wind generating set corresponding to the first wind speed derivative variable and the second wind speed derivative variable; the state information is used for representing whether each wind generating set breaks down or not;
the second logistic regression model establishing module is used for establishing a logistic regression model according to the first wind speed derivative variable, the second wind speed derivative variable and the state information of each wind generating set;
and the second prediction module is used for predicting the fault probability of the wind generating set to be predicted based on the established logistic regression model and the first wind speed derivative variable and the second wind speed derivative variable of the wind generating set to be predicted.
11. The apparatus according to claim 10, wherein the second logistic regression model establishing module is specifically configured to perform fitting calculation on the first wind speed derivative variable, the second wind speed derivative variable, and the state information of each wind turbine generator set to obtain modeling parameters of the logistic regression model;
and establishing a corresponding relation by taking the first wind speed derivative variable and the second wind speed derivative variable as independent variables and taking the fault probability of the wind generating set as dependent variables based on the modeling parameters to generate the logistic regression model.
12. The device according to claim 11, wherein the second prediction module is specifically configured to obtain a first wind speed derivative variable and a second wind speed derivative variable of the wind turbine generator set to be predicted;
and inputting the first wind speed derivative variable and the second wind speed derivative variable of the wind generating set to be predicted into the logistic regression model, and obtaining the fault probability of the wind generating set to be predicted according to the established corresponding relation which takes the first wind speed derivative variable and the second wind speed derivative variable as independent variables and takes the fault probability of the wind generating set as dependent variables.
13. A computer-readable storage medium having stored thereon computer program instructions, which when executed by a processor, implement the method of any one of claims 1-6.
14. A computer apparatus, the apparatus comprising:
at least one processor, at least one memory, and computer program instructions stored in the memory,
the computer program instructions, when executed by the processor, implement the method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910004963.1A CN111396265B (en) | 2019-01-03 | 2019-01-03 | Fault prediction method and device of wind generating set |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910004963.1A CN111396265B (en) | 2019-01-03 | 2019-01-03 | Fault prediction method and device of wind generating set |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111396265A CN111396265A (en) | 2020-07-10 |
| CN111396265B true CN111396265B (en) | 2023-02-03 |
Family
ID=71435835
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910004963.1A Active CN111396265B (en) | 2019-01-03 | 2019-01-03 | Fault prediction method and device of wind generating set |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111396265B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112149903B (en) * | 2020-09-22 | 2022-04-08 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Primary frequency modulation analysis and optimization method of thermal power generating unit based on BP neural network algorithm |
| CN115455368B (en) * | 2022-11-09 | 2023-02-03 | 中国人民解放军国防科技大学 | Troposphere lower-layer gust calculation method based on turbulence intensity and average wind speed |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102006001613A1 (en) * | 2006-01-11 | 2007-07-12 | Repower Systems Ag | Method for operating a wind turbine and wind turbine |
| KR20100069045A (en) * | 2008-12-16 | 2010-06-24 | 노영규 | Variable power generator for wind power generation |
| CN101852172A (en) * | 2010-03-09 | 2010-10-06 | 山东科技大学 | Method for calculating input wind speed of wind generating sets according to wake effect in wind power station |
| CN103026266A (en) * | 2010-06-21 | 2013-04-03 | 维斯塔斯风力系统集团公司 | Estimation of wind conditions at a wind turbine |
| CN103174590A (en) * | 2013-04-03 | 2013-06-26 | 无锡韦伯风能技术有限公司 | Adjusting method of medium-sized wind power machine unit power regulator |
| CN103400302A (en) * | 2013-07-01 | 2013-11-20 | 国家电网公司 | Risk perception and early warning method and system for cascading failures of wind power base |
| WO2015108154A1 (en) * | 2014-01-17 | 2015-07-23 | 国立大学法人京都大学 | Maximum peak-gust wind speed prediction system, computer program, and wind-gust warning information announcement system |
| CN106529700A (en) * | 2016-10-20 | 2017-03-22 | 华北电力大学（保定） | Space-time correlation based construction method and system of wind power plant prediction model |
| WO2018008571A1 (en) * | 2016-07-04 | 2018-01-11 | 株式会社日本製鋼所 | Wind turbine monitoring device, wind turbine monitoring method, wind turbine monitoring program, and storage medium |
| CN107781105A (en) * | 2016-08-31 | 2018-03-09 | 北京天诚同创电气有限公司 | The pitch control method and device of wind-driven generator |
| CN108223268A (en) * | 2016-12-14 | 2018-06-29 | 北京金风科创风电设备有限公司 | The method for controlling number of revolution and device of wind power generating set |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DK2251543T3 (en) * | 2009-05-14 | 2017-02-27 | Alstom Renewable Technologies | Method and system for predicting the occurrence of a gust at a wind turbine |
| EP2705252B1 (en) * | 2011-05-06 | 2017-03-22 | Vestas Wind Systems A/S | Method and apparatus for protecting wind turbines from extreme events |
| CN104533713B (en) * | 2014-11-03 | 2017-05-17 | 新疆金风科技股份有限公司 | Method and device for controlling wind generating set to be shut down |
| CN108223269B (en) * | 2016-12-14 | 2019-07-23 | 北京金风科创风电设备有限公司 | Method and device for passing through overspeed fault of wind generating set |
| CN106972549B (en) * | 2017-05-12 | 2019-11-12 | 北京金风科创风电设备有限公司 | Method and device for energy management of a wind farm |
-
2019
- 2019-01-03 CN CN201910004963.1A patent/CN111396265B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102006001613A1 (en) * | 2006-01-11 | 2007-07-12 | Repower Systems Ag | Method for operating a wind turbine and wind turbine |
| KR20100069045A (en) * | 2008-12-16 | 2010-06-24 | 노영규 | Variable power generator for wind power generation |
| CN101852172A (en) * | 2010-03-09 | 2010-10-06 | 山东科技大学 | Method for calculating input wind speed of wind generating sets according to wake effect in wind power station |
| CN103026266A (en) * | 2010-06-21 | 2013-04-03 | 维斯塔斯风力系统集团公司 | Estimation of wind conditions at a wind turbine |
| CN103174590A (en) * | 2013-04-03 | 2013-06-26 | 无锡韦伯风能技术有限公司 | Adjusting method of medium-sized wind power machine unit power regulator |
| CN103400302A (en) * | 2013-07-01 | 2013-11-20 | 国家电网公司 | Risk perception and early warning method and system for cascading failures of wind power base |
| WO2015108154A1 (en) * | 2014-01-17 | 2015-07-23 | 国立大学法人京都大学 | Maximum peak-gust wind speed prediction system, computer program, and wind-gust warning information announcement system |
| WO2018008571A1 (en) * | 2016-07-04 | 2018-01-11 | 株式会社日本製鋼所 | Wind turbine monitoring device, wind turbine monitoring method, wind turbine monitoring program, and storage medium |
| CN107781105A (en) * | 2016-08-31 | 2018-03-09 | 北京天诚同创电气有限公司 | The pitch control method and device of wind-driven generator |
| CN106529700A (en) * | 2016-10-20 | 2017-03-22 | 华北电力大学（保定） | Space-time correlation based construction method and system of wind power plant prediction model |
| CN108223268A (en) * | 2016-12-14 | 2018-06-29 | 北京金风科创风电设备有限公司 | The method for controlling number of revolution and device of wind power generating set |
Non-Patent Citations (3)
| Title |
|---|
| 台风环境下考虑地理高程信息的输电通道结构失效故障概率评估方法;陈莹等;《电网技术》;20170929(第07期);全文 * |
| 针对金风750kW系列风力发电机组频繁起停问题的分析与讨论;陈锋;《电气制造》;20110525(第05期);全文 * |
| 风力发电机组载荷计算中的极端风况分析;李媛等;《沈阳工业大学学报》;20080815;第30卷(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111396265A (en) | 2020-07-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2020399503B2 (en) | Method and apparatus for detecting yaw-to-wind abnormality, and device and storage medium thereof | |
| US20190072075A1 (en) | Method and device for detecting equivalent load of wind turbine | |
| CA3162702A1 (en) | Method and apparatus for detecting fault, method and apparatus for training model, and device and storage medium | |
| CN111396265B (en) | Fault prediction method and device of wind generating set | |
| CN110365059B (en) | Optical power prediction method and device | |
| CN104252649A (en) | Regional wind power output prediction method based on correlation between multiple wind power plants | |
| CN112855408B (en) | Early warning method and early warning device for hydroelectric generating set | |
| CN107305651B (en) | Power transmission system reliability assessment method and system | |
| DK2981711T3 (en) | PROCEDURE FOR EFFICIENCY MONITORING OF A WINDOW ENERGY PARK | |
| CN111311021A (en) | Theoretical power prediction method, device, equipment and storage medium for wind power plant | |
| CN105356462A (en) | Wind farm harmonic evaluation prediction method | |
| CN111401596B (en) | Method and device for generating wind speed index | |
| CN116090216A (en) | Power generation power prediction method and system based on typical correlation analysis | |
| CN115577854A (en) | Quantile regression wind speed interval prediction method based on EEMD-RBF combination | |
| CN111350638A (en) | Method and device for calculating power generation loss of wind generating set | |
| CN110852544B (en) | Reliability evaluation method and device for wind generating set | |
| Veena et al. | Parametric models for predicting the performance of wind turbines | |
| De Klerk et al. | Power calculation accuracy as a function of wind data resolution | |
| CN116757101B (en) | Cabin wind speed correction method and system based on mechanism model and neural network | |
| Zhengganzhe et al. | Risk forecast assessment of transmission line capacity increase under cyberattacks | |
| Ábrahám et al. | Unsicherheitsquantifizierung von Windparkmodellen Uncertainty quantification of wind farm models | |
| CN117807457A (en) | Data detection method and device and electronic equipment | |
| Smith et al. | Evaluation of Synthetic Wind Speed Time Series for Reliability Analysis of Offshore Wind Farms | |
| Castellani et al. | Advanced Methods for Wind Turbine Performance Analysis Based on SCADA Data and CFD Simulations. Energies 2023, 16, 1081 | |
| CN117091802A (en) | Calibration method of wind speed turbulence model based on measured data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 830026 No. 107, Shanghai Road, Urumqi economic and Technological Development Zone, the Xinjiang Uygur Autonomous Region Patentee after: Jinfeng Technology Co.,Ltd. Address before: 830026 No. 107, Shanghai Road, Urumqi economic and Technological Development Zone, the Xinjiang Uygur Autonomous Region Patentee before: XINJIANG GOLDWIND SCIENCE & TECHNOLOGY Co.,Ltd. |