A method for predicting the failure rate of wind turbine gearbox based on different operating conditions

Abstract

The present invention discloses a method for predicting the failure rate of a fan gearbox based on different operating states, which is characterized by comprising the following steps: Step 1: Combined with the "bathtub curve" in the Weibull fault distribution model, the operating state s is added to the Weibull fault analysis model, and the new fault analysis model λ(t, s) is considered together with the time factor t to construct a new fault analysis model λ(t, s); λ(t, s) is expressed as four cases; the expression of λ(t, s) in the four cases is determined; the four cases in step 2 are combined and analyzed to construct a gearbox fault analysis combination model λ(t, s1, s2, s3, s4). The present invention adds a new conditional factor, the gearbox operating state, to the previous prediction of the failure rate of the fan gearbox based on the Weibull fault distribution, taking into account the time factor, and derives the probability of predicting the failure of a specific number of fan gearboxes at a specific time under the combination of the four different operating states of the fan.

Description

Technical Field

[0001] The present invention relates to the technical field of wind turbine fault analysis, and in particular to a method for predicting the failure rate of a wind turbine gearbox based on different operating states. Background Art

[0002] Wind power generation is an important part of renewable energy generation. Its core component, the gearbox, often has a high failure rate due to factors such as operating time and operating status. Its failure frequency ranks fourth among all major equipment, second only to the inverter, generator and pitch system. The downtime caused by gearbox failure accounts for 19.4% of the total downtime caused by various reasons. At the same time, the complex maintenance process of wind turbine gearbox leads to relatively high maintenance costs and maintenance time. Therefore, it is necessary to study the reliability of wind turbine gearbox.

[0003] Traditional technologies for predicting the failure rate of wind turbine gearboxes all study the impact of time factors on the failure rate of wind turbine gearboxes based on the Weibull fault distribution curve. However, the reliability of this single-factor fault analysis needs to be improved, especially for offshore wind turbines, which are often worse in operating condition than those on land due to seawater corrosion. Therefore, it is particularly important to refer to the operating condition when conducting fault prediction analysis. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the problems existing in the above-mentioned existing methods for predicting the failure rate of wind turbine gearboxes, the present invention is proposed.

[0006] Therefore, an object of the present invention is to provide a method for predicting the failure rate of a wind turbine gearbox based on different operating conditions, in order to consider the influence of the wind turbine operating condition on the failure rate of the wind turbine gearbox.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for predicting the failure rate of a wind turbine gearbox based on different operating states, characterized in that it comprises the following steps:

[0008] Step 1: Combined with the "bathtub curve" in the Weibull fault distribution model, the operating state s is added to the Weibull fault analysis model, and the time factor t is considered together to construct a new fault analysis model λ(t,s);

[0009] Step 2: Express λ(t,s) into four situations: an increase in the overall failure rate, a shortened overall life span and an earlier failure time, an accelerated failure rate during the failure period, and a shortened life span during the failure period;

[0010] Step 3: Determine the λ(t,s) expression in four cases, and further calculate and derive the reliability function, failure cumulative distribution function, and failure probability density function of the four cases respectively;

[0011] Step 4: Perform a combined analysis on the four situations in step 2, construct a gearbox fault analysis combined model λ(t,s1,s2,s3,s4), determine the expression and calculate the reliability function, cumulative distribution function, failure probability density function and the probability of failure of a specific number of devices at a specific time of the combined analysis model.

[0012] As a preferred solution of the method for predicting the failure rate of a wind turbine gearbox based on different operating states of the present invention, the increase in the overall failure rate in step 2 is specifically manifested as follows: under the premise of the same time t, a wind turbine gearbox in a poor state has a higher failure rate than a wind turbine gearbox in a good state;

[0013] Among them, the shortened overall life and early failure time are specifically manifested as follows: the fan gearbox with poor operating status can reach the same number of failures in a shorter time than the fan gearbox with good operating status;

[0014] Among them, the accelerated failure rate during the failure period is specifically manifested as follows: the failure rate of the wind turbine gearbox in a poor operating state is faster than that of the wind turbine gearbox in a good operating state;

[0015] Among them, the shortened life during the failure period is specifically manifested as follows: the wind turbine gearbox in poor operating condition will experience wear and tear earlier than the wind turbine gearbox in good operating condition, and its life will be shortened.

[0016] As a preferred solution of the method for predicting the failure rate of a wind turbine gearbox based on different operating states according to the present invention, the failure rate model expression corresponding to the increase in the overall failure rate is:

[0017] The failure rate model expression corresponding to the overall life shortening and failure time advancement is:

[0018] The failure rate model expression corresponding to the accelerated failure rate during the failure period is:

[0019] The failure rate model expression corresponding to the shortened failure period life is:

[0020] Among them, β is the shape parameter, θ is the scale parameter, t is the fault occurrence time, and s is the score value of the wind turbine historical operation data.

[0021] As a preferred solution of the method for predicting the failure rate of a wind turbine gearbox based on different operating states according to the present invention, the expression of λ(t, s1, s2, s3, s4) in step 4 is:

[0022]

[0023] Among them, S1 represents the influence of the fan operating state on the failure rate of the gearbox during the failure period, satisfying S1∈(0,+∞), S2 represents the influence of the fan operating state on the life of the gearbox during the failure period, satisfying S2∈(0,+∞), S3 represents the influence of the fan operating state on the overall life of the gearbox, satisfying S3∈[0,+∞). S4 represents the influence of the fan operating state on the overall failure rate of the gearbox, satisfying S4∈[0,1).

[0024] As a preferred solution of the method for predicting the failure rate of a wind turbine gearbox based on different operating states according to the present invention, wherein: R(t, s1, s2, s3, s4), f(t, s1, s2, s3, s4) and F(t, s1, s2, s3, s4) corresponding functions are derived respectively according to λ(t, s1, s2, s3, s4);

[0025] Among them, R(t,s1,s2,s3,s4) represents the probability function of a single wind turbine having no failure within t years, F(t,s1,s2,s3,s4) represents the probability function of a single wind turbine having failure within t years, and f(t,s1,s2,s3,s4) represents the failure probability density function.

[0026] Beneficial effects of the invention: The invention adds a new conditional factor, the gearbox operating state, to the previous prediction of the wind turbine gearbox failure rate based on Weibull fault distribution, which takes into account the time factor. The model's "reliability function", "fault cumulative distribution function", "fault probability density function" and "probability of failure of a specific number of devices at a specific time" are derived considering the combination of four different operating states of the wind turbine. This helps wind power companies to control operation and maintenance control costs and inventory costs within a reasonable range, thereby helping to improve the stability of wind power generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0028] Figure 1 A schematic flow chart of a method for predicting a failure rate of a wind turbine gearbox based on different operating conditions proposed by the present invention. DETAILED DESCRIPTION

[0029] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0030] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0031] Example 1

[0032] Reference Figure 1 , as an embodiment of the present invention, provides a method for predicting the failure rate of a wind turbine gearbox based on different operating states, the method comprising the following steps:

[0033] Step 1: Combined with the "bathtub curve" in the Weibull fault distribution model, the operating state s is added to the Weibull fault analysis model, and the time factor t is considered together to construct a new fault analysis model λ(t,s);

[0034] Step 2: Express λ(t,s) as four situations, namely: increased overall failure rate, shortened overall life and early failure time, accelerated failure rate during failure period and shortened life during failure period. Specifically, the increased overall failure rate is manifested as follows: under the same time t, the wind turbine gearbox in poor condition has a higher failure rate than the wind turbine gearbox in good condition; among them, the shortened overall life and early failure time are specifically manifested as follows: the wind turbine gearbox in poor operating condition can reach the same number of failures in a shorter time than the wind turbine gearbox in good operating condition; among them, the accelerated failure rate during failure period is specifically manifested as follows: the wind turbine gearbox in poor operating condition has a faster failure rate than the wind turbine gearbox in good operating condition; among them, the shortened life during failure period is specifically manifested as follows: the wear failure period of the wind turbine gearbox in poor operating condition will arrive earlier than that of the wind turbine gearbox in good operating condition, and the life is shortened.

[0035] Step 3: Determine the λ(t,s) expression for the four cases, and further calculate and derive the reliability function, failure cumulative distribution function, and failure probability density function for the four cases.

[0036] Specifically, the failure rate model expression corresponding to the increase in the overall failure rate is as follows: And further deduced: Get the corresponding functions of R(t,s), f(t,s) and F(t,s):

[0037]

[0038]

[0039]

[0040] The failure rate model expression corresponding to the overall life shortening and failure time advancing is: And further derive the corresponding functions of R(t,s), f(t,s) and F(t,s):

[0041]

[0042]

[0043]

[0044] The failure rate model expression corresponding to the accelerated failure rate during the failure period is: Further deduction, the corresponding functions of R(t,s), f(t,s) and F(t,s) are:

[0045]

[0046]

[0047]

[0048] The failure rate model expression corresponding to the shortened failure period life is: Further deduction:

[0049]

[0050]

[0051]

[0052] Among them, β is the shape parameter, θ is the scale parameter, t is the time when the fault occurs, s is the score value of the wind turbine historical operation data, R(t,s) corresponds to the reliability function, F(t,s) corresponds to the cumulative distribution function of the fault, and f(t,s) is the fault probability density function.

[0053] Step 4: Perform a combined analysis on the four situations in step 2, construct a gearbox fault analysis combined model λ(t,s1,s2,s3,s4), determine the expression and calculate the reliability function, cumulative distribution function, failure probability density function and the probability of failure of a specific number of devices at a specific time of the combined analysis model.

[0054] According to λ(t,s1,s2,s3,s4), derive the corresponding functions of R(t,s1,s2,s3,s4), f(t,s1,s2,s3,s4) and F(t,s1,s2,s3,s4);

[0055] Among them, R(t,s1,s2,s3,s4) represents the probability function of no failure of a single wind turbine within t years, that is, the reliability function; F(t,s1,s2,s3,s4) represents the probability function of failure of a single wind turbine within t years, that is, the cumulative distribution function of failure; and f(t,s1,s2,s3,s4) represents the failure probability density function.

[0056] For further analysis, the expression of λ(t,s1,s2,s3,s4) in step 4 is:

[0057]

[0058] Among them, S1 represents the influence of the fan operating state on the failure rate of the gearbox during the failure period, S2 represents the influence of the fan operating state on the life of the gearbox during the failure period, S3 represents the influence of the fan operating state on the overall life of the gearbox, and S4 represents the influence of the fan operating state on the overall failure rate of the gearbox, satisfying S1∈(0,+∞), S2∈(0,+∞), S3∈[0,+∞), S4∈[0,1),

[0059] Further deduction, the corresponding functions of R(t,s1,s2,s3,s4), f(t,s1,s2,s3,s4) and F(t,s1,s2,s3,s4) are:

[0060]

[0061]

[0062]

[0063] Assume that n wind turbine gearboxes have different operating states in the t-th year, and the probability of the i-th wind turbine gearbox failing in the t-th year is denoted as λ i (t, s i ), N i )(t) = 1 indicates that wind turbine i has failed within t years, and N i (t) = 0 indicates that wind turbine i has not failed within t years, and s ij represents the performance of wind turbine i in the s j state, where i ∈ [1, n] and j = 1, 2, 3, 4, so we have

[0064] Make a specific analysis based on whether the operating states of the wind turbine gearboxes in the t-th year are the same or different;

[0065] 1) When the operating states of all wind turbine gearboxes in the t-th year are the same, the probability P(N(t, s1, s2, s3, s4) = x) of x failures occurring in the gearbox in the t-th year follows a binomial distribution, so we have:

[0066]

[0067] 2) When the operating states of the wind turbine gearboxes in the t-th year are different, the probability of the i-th wind turbine gearbox failing in the t-th year conforms to a 0-1 distribution, that is, N i (t) ~ B(1, p i ), and the probability of x failures occurring in the gearbox in the t-th year is:

[0068] ① When x = 1, we have:

[0069]

[0070] ② When 1 < x < n, we have:

[0071]

[0072] ③ When x = n, we have:

[0073]

[0074] Based on the previous prediction of the failure rate of wind turbine gearboxes based on the Weibull failure distribution considering time factors, this invention adds a new conditional factor of the operating state of the gearbox, and deduces the "reliability function", "cumulative failure distribution function", "failure probability density function" of the model, and the "probability of a specific number of devices failing at a specific time" under the combination of 4 different operating states of the wind turbine, which helps wind power generation enterprises control the operation and maintenance control cost and inventory cost within a reasonable range, thus contributing to improving the stability of wind power generation.

[0075] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A method for predicting the failure rate of a wind turbine gearbox based on different operating conditions, characterized in that: The following steps are involved: Step 1: Combined with the "bathtub curve" in the Weibull fault distribution model, the operating state s is added to the Weibull fault analysis model, and the time factor t is considered together to construct a new fault analysis model λ(t,s); Step 2: Express λ(t,s) into four situations: increased overall failure rate, shortened overall life and earlier failure time, faster failure rate during failure period, and shortened life during failure period; Step 3: Determine the λ(t,s) expression in four cases, and further calculate and derive the reliability function, failure cumulative distribution function, and failure probability density function of the four cases respectively; Step 4: Perform a combined analysis on the four situations in step 2, construct a gearbox fault analysis combined model λ(t, s1, s2, s3, s4), determine the expression and calculate the reliability function, cumulative distribution function, fault probability density function and the probability of a specific number of devices failing at a specific time of the combined model; Among them, the increase of the overall failure rate in the step 2 is specifically manifested as follows: under the premise of the same time t, the fan gearbox in a poor state has a higher failure rate than the fan gearbox in a good state; the overall life is shortened and the failure time is advanced, which is specifically manifested as follows: the fan gearbox in a poor operating state can reach the same number of failures in a shorter time than the fan gearbox in a good operating state; the failure rate during the failure period is specifically manifested as follows: the failure rate of the fan gearbox in a poor operating state is faster than that of the fan gearbox in a good operating state; the shortened life during the failure period is specifically manifested as follows: the wear failure period of the fan gearbox in a poor operating state will arrive earlier than that of the fan gearbox in a good operating state, and the life is shortened; The failure rate model expression corresponding to the increase in the overall failure rate is: The failure rate model expression corresponding to the overall life shortening and failure time advancement is: The failure rate model expression corresponding to the accelerated failure rate during the failure period is: The failure rate model expression corresponding to the shortened failure period life is: Among them, β is the shape parameter, θ is the scale parameter, t is the fault occurrence time, and s is the score value of the wind turbine historical operation data; The expression of λ(t,s1,s2,s3,s4) in step 4 is: Among them, S1 represents the influence of the fan operating state on the failure rate of the gearbox during the failure period, satisfying S1∈(0,+∞); S2 represents the influence of the fan operating state on the life of the gearbox during the failure period, satisfying S2∈(0,+∞); S3 represents the influence of the fan operating state on the overall life of the gearbox, satisfying S3∈[0,+∞); S4 represents the influence of the fan operating state on the overall failure rate of the gearbox, satisfying S4∈[0,1); According to λ(t,s1,s2,s3,s4), derive the corresponding functions of R(t,s1,s2,s3,s4), f(t,s1,s2,s3,s4) and F(t,s1,s2,s3,s4); Among them, R(t,s1,s2,s3,s4) represents the probability function of a single wind turbine having no failure within t years, F(t,s1,s2,s3,s4) represents the probability function of a single wind turbine having failure within t years, and f(t,s1,s2,s3,s4) represents the failure probability density function.

Images