A reliability assessment method for wind turbines based on TS fault tree

Abstract

The present invention provides a reliability assessment method for a wind turbine generator set based on a T-S fault tree, comprising: constructing a first T-S dynamic fault tree model according to a unit step function; obtaining a first fault probability density function according to the first T-S dynamic fault tree model; constructing a second T-S dynamic fault tree model according to the input and output rules of a multidimensional T-S dynamic gate based on the first T-S dynamic fault tree model; obtaining a first probability distribution function according to the second T-S dynamic fault tree model based on the first fault probability density function; constructing a third T-S dynamic fault tree model by introducing input and output rules under multi-state influence based on the second T-S dynamic fault tree model; obtaining a second fault probability distribution function according to the third T-S dynamic fault tree model based on the first probability distribution function; obtaining the importance of each event and evaluating the wind turbine generator set to be tested according to the second fault probability distribution function. The present invention solves the problem that the T-S fault tree has errors and cumbersome calculations.

Description

Technical Field

[0001] The invention relates to the field of electric power technology, and in particular to a reliability assessment method for a wind turbine generator set based on a TS fault tree. Background Art

[0002] The failure process of an actual wind turbine system is usually affected by multiple factors including wind speed, weather, temperature, and the life of each component. These influences are dynamic, random, volatile, and time-sensitive. The existing TS fault tree can characterize static and dynamic fault behaviors and describe the static and dynamic characteristics of the wind turbine system, but it is limited by the binary event relationship between working and shutting down, lacks research on intermediate states, cannot describe the polymorphic characteristics of the system, and has too little reliability data for analyzing the probability of failure. At the same time, it also lacks consideration of non-time factors that affect reliability. At present, the number of divisions of the task time is infinitely increased to make it close to a continuous system. This method still has errors and the calculation is cumbersome. Summary of the invention

[0003] In order to overcome the deficiencies of the prior art, the purpose of the present invention is to provide a reliability assessment method for a wind turbine generator set based on a TS fault tree, which solves the problem in the prior art that the TS fault tree has errors and cumbersome calculations.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] A reliability assessment method for wind turbines based on TS fault tree, comprising:

[0006] Determine the relationship between various events of the wind turbine system to be tested and the unit faults;

[0007] Based on the relationship between various levels of events of the wind turbine system to be tested and the unit fault, a first TS dynamic fault tree model is constructed according to a unit step function; the various levels of events of the wind turbine system to be tested include: lower level events, upper level events and top events;

[0008] According to the first TS dynamic fault tree model, a first fault probability density function is obtained; the first fault probability density function includes: a first lower-level event fault probability density function and a first upper-level event fault probability density function;

[0009] Based on the first TS dynamic fault tree model, a second TS dynamic fault tree model is constructed according to the input and output rules of the multi-dimensional TS dynamic gate;

[0010] Based on the first fault probability density function, according to the second TS dynamic fault tree model, a first probability distribution function is obtained;

[0011] Based on the second TS dynamic fault tree model, the input and output rules under the influence of polymorphism are introduced to construct a third TS dynamic fault tree model;

[0012] Based on the first probability distribution function, according to the third TS dynamic fault tree model, a second fault probability distribution function is obtained; the second fault probability distribution function is a top event fault probability distribution function;

[0013] According to the second fault probability distribution function, the importance of each event is obtained;

[0014] The wind turbine set to be tested is evaluated according to the importance.

[0015] Preferably, the first probability distribution function comprises:

[0016] A first lower-level event probability distribution function and a first upper-level event probability distribution function.

[0017] Preferably, the obtaining of the first probability distribution function based on the first fault probability density function and according to the second TS dynamic fault tree model comprises:

[0018] Based on the first lower-level event failure probability density function, according to the input rule of the multi-dimensional TS dynamic gate, a second lower-level event failure probability density function is obtained;

[0019] Obtaining a first subordinate event probability distribution function according to the second subordinate event failure probability density function;

[0020] Based on the first upper-level event failure probability density function, according to the output rule of the multi-dimensional TS dynamic gate, the second upper-level event failure probability density function is obtained;

[0021] The first superior event probability distribution function is obtained according to the second superior event failure probability density function.

[0022] Preferably, the obtaining of the second fault probability distribution function based on the first probability distribution function and according to the third TS dynamic fault tree model comprises:

[0023] Based on the first subordinate event probability distribution function, according to the input rule under the influence of polymorphism, a second subordinate event probability distribution function is obtained;

[0024] Based on the first upper-level event probability distribution function, according to the output rule under the influence of polymorphism, a second upper-level event probability distribution function is obtained;

[0025] According to the second lower-level event probability distribution function and the second upper-level event probability distribution function, a top-level event failure probability distribution function is obtained.

[0026] Preferably, the importance of each event includes:

[0027] Probability importance, key importance, FV importance and comprehensive importance.

[0028] Preferably, the first higher-level event failure probability density function expression is:

[0029]

[0030] Among them, f i (t i ) is the probability density function of the first lower-level event failure, f i (t y ) is the probability density function of the first upper-level event failure, i is the failure time of the lower-level event, t i is the failure time of the ith subordinate event, t y The upper-level event occurs at the failure time of the lower-level event, f i is the failure probability density function.

[0031] Preferably, the expression of the first subordinate event probability distribution function is:

[0032]

[0033] Where k is the number of image factors, h K represents the Kth influencing factor, F i (h ρ ) is h ρ Subordinate events under the influence of factor X i The failure probability distribution function.

[0034] Preferably, the expression of the first superior event probability distribution function is:

[0035]

[0036] Preferably, the expression of the second subordinate event probability distribution function is:

[0037]

[0038] in, The status of the subordinate event is represented by becomes The probability of all possible paths, S represents the state of the event; The status of the subordinate event is represented by becomes In becoming The probability of all possible paths, A is a certain state.

[0039] Preferably, the expression of the second superior event probability distribution function is:

[0040]

[0041] in, The state of the parent event y is determined by becomes The failure probability density function, q is the qth influencing factor, s y is a state of the top event, b y The probability function of the fault state occurrence of the superior event.

[0042] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0043] The present invention provides a reliability assessment method for a wind turbine generator set based on a TS fault tree, comprising: determining the mutual relationship between events at various levels of a wind turbine generator set system to be tested and the unit fault; based on the mutual relationship between events at various levels of the wind turbine generator set system to be tested and the unit fault, constructing a first TS dynamic fault tree model according to a unit step function; the events at various levels of the wind turbine generator set system to be tested include: lower-level events, upper-level events and top events; according to the first TS dynamic fault tree model, obtaining a first fault probability density function; the first fault probability density function includes: a first lower-level event fault probability density function and a first upper-level event fault probability density function; based on the first TS dynamic fault tree model, constructing a second TS dynamic fault tree model according to the input and output rules of a multidimensional TS dynamic gate; based on the first A fault probability density function, according to the second TS dynamic fault tree model, a first probability distribution function is obtained; based on the second TS dynamic fault tree model, the input and output rules under the influence of multiple states are introduced to construct a third TS dynamic fault tree model; based on the first probability distribution function, according to the third TS dynamic fault tree model, a second fault probability distribution function is obtained; the second fault probability distribution function is a top event fault probability distribution function; according to the second fault probability distribution function, the importance of each event is obtained; according to the importance, the wind turbine to be tested is evaluated. The present invention can characterize any dynamic failure behavior through TS dynamic gates and continuous time description rules, and an algorithm is given to solve the top event fault probability, which makes up for the shortcomings of the dynamic fault tree in terms of description means and calculation methods. In addition, the input and output rules of the TS dynamic fault tree under multiple external influencing factors and the three states of wind turbine shutdown, frequency reduction and normal operation are introduced, and thus the conversion probability distribution function of the wind turbine in each state is obtained, and the conversion probability distribution function of the wind turbine in each state under multiple influencing factors is obtained, which makes up for the diversification and multi-angle of reliability evaluation. Finally, the probability distribution function of wind turbine failure is used to obtain four reliability indices of probability importance, critical importance, FV importance and comprehensive importance to evaluate the impact of the part on the whole and to reflect the impact of each subsystem of the wind turbine on the wind turbine failure. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. 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 paying creative labor.

[0045] Figure 1 A flow chart of a reliability assessment method for a wind turbine generator system based on a TS fault tree provided by an embodiment of the present invention;

[0046] Figure 2 A detailed flow chart of a reliability assessment method for a wind turbine generator system based on a TS fault tree provided in an embodiment of the present invention;

[0047] Figure 3 A schematic diagram of a fault tree of a wind turbine generator system provided by an embodiment of the present invention;

[0048] Figure 4 A schematic diagram of a first TS dynamic fault tree model provided by an embodiment of the present invention;

[0049] Figure 5 A schematic diagram of importance provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] The purpose of the present invention is to provide a reliability assessment method for a wind turbine generator set based on a TS fault tree, which solves the problem in the prior art that the TS fault tree has errors and is cumbersome in calculation.

[0052] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0053] like Figure 1-2 As shown, the present invention provides a reliability assessment method for a wind turbine generator set based on a TS fault tree, comprising:

[0054] Step 100: Determine the relationship between various levels of events and unit faults of the wind turbine system to be tested;

[0055] Specifically, Figure 3 As shown in the figure, wind turbine failures are divided into six key subsystem failures according to their functions, namely, the rotor system, pitch system, generator, yaw system, wind measurement system, nacelle base and tower. The nacelle base and tower provide physical support for the wind turbine; the rotor system and permanent magnet synchronous generator realize the transmission and conversion of wind energy; the pitch system and yaw system adjust the blade pitch angle and the wind direction of the wind turbine according to the wind speed and wind direction measured by the wind measurement system to control the operating status of the wind turbine.

[0056] Table 1 is a basic event table, which is as follows:

[0057] Table 1 Basic event table

[0058]

[0059] Step 200: Based on the relationship between various levels of events of the wind turbine system to be tested and the unit fault, a first TS dynamic fault tree model is constructed according to a unit step function; the various levels of events of the wind turbine system to be tested include: lower level events, upper level events and top events;

[0060] Specifically, Figure 4 As shown, the unit step function is used to express the event occurrence timing dependency of different lower-level events of the TS dynamic gate, that is:

[0061]

[0062] Where: t a ,t b are any two different subordinate events X among n subordinate events. a , X b (a=1, 2, ..., n; b=1, 2, ..., n); u(t b -t a )、u(t a -t b ) is a unit step function, representing the lower level event X a Before subordinate event X b Occurs, lower level event X b Before subordinate event X a The amplitude of the unit step function is 1, and its value will not change when the fault probability density function of the parent event y is subsequently solved.

[0063] Let the order rule R = [R(1), R(2), ..., R(l), ..., R(r)], where the order rule R (l) (l=1,2,...,r,r=n!) is composed of a set of unit step functions, which represents the order of the fault moments of n subordinate events, that is, the formula:

[0064]

[0065] Where: n is the number of lower-level events; t1, t2, ..., t i-1 , t i , ..., t n-2 , t n-1 , t n They are the failure moments of the 1st, 2nd, 3rd, ..., n-2, n-1, and nth subordinate events respectively.

[0066] Using the unit pulse function δ(t i -ty ) to describe the failure moment of the superior event, that is, the formula:

[0067]

[0068] Where: t y is the failure time of the superior event y; t i For the subordinate event x i The fault time t of (i=1, 2, 3, ..., n) i When the superior event occurs y =t i , indicating that the superior event is in the subordinate event x i The failure time t i Occurs, at which time the unit pulse function δ(t i -t y )=∞; when the superior event does not occur, t y ≠t i , at this time the unit pulse function δ(t i -t y )=0;

[0069] Using the unit pulse function screening properties, the fault probability density function f of the upper-level event y can be constructed i (t y )(First-level event failure probability density function).

[0070]

[0071] Where: f i (t i ) is the subordinate event X i The failure probability density function (the first lower-level event failure probability density function) is

[0072] (1) The natural number o(ti) (i=1, 2,"n-1, n) in the input rule represents the occurrence sequence of the subordinate event xi (i=1, 2,"n-1, n), that is, the size relationship of the subordinate event failure time t1, t2, ..., tn, where the smaller value occurs before the larger value.

[0073] (2) The unit pulse function δ(ti-ty) in the output rule can describe the fault moment of the upper-level event y. Table 2 is the event sequence description rule table of the TS dynamic gate, which is as follows:

[0074] Table 2 TS dynamic gate event sequence description rules

[0075]

[0076] Each row in the table represents a rule. For example, rule 1: Input the subordinate events X1 to X n Failure in the order of 1, 2, 3, ..., n-1, n, i.e., t1<t2<t3<…<t n-1 <t n , the order rules at this time.

[0077] O (2) =u(t n -t n-1 )u(t n-1 -t n-2 )…u(t3-t1);

[0078] And the unit pulse function representing the fault moment of the superior event y is f(1)(ty)

[0079] In short, TS dynamic gate can describe any form of static and dynamic event relationship. Moreover, TS dynamic gate is described by event sequence description rules, which reduces the complexity of calculation and is easy to program.

[0080] Step 300: obtaining a first fault probability density function according to the first TS dynamic fault tree model; the first fault probability density function includes: a first lower-level event fault probability density function and a first upper-level event fault probability density function;

[0081] Step 400: Based on the first TS dynamic fault tree model, a second TS dynamic fault tree model is constructed according to the input and output rules of the multi-dimensional TS dynamic gate;

[0082] Step 500: Based on the first fault probability density function and the second TS dynamic fault tree model, a first probability distribution function is obtained;

[0083] Step 600: Based on the second TS dynamic fault tree model, the input and output rules under the influence of multiple states are introduced to construct a third TS dynamic fault tree model;

[0084] Step 700: Based on the first probability distribution function and according to the third TS dynamic fault tree model, a second fault probability distribution function is obtained; the second fault probability distribution function is a top event fault probability distribution function;

[0085] Step 800: Obtaining the importance of each event according to the second fault probability distribution function;

[0086] Step 900: Evaluate the wind turbine group to be tested according to the importance.

[0087] Furthermore, the first probability distribution function includes:

[0088] A first lower-level event probability distribution function and a first upper-level event probability distribution function.

[0089] Further, the first probability distribution function is obtained based on the first fault probability density function and according to the second TS dynamic fault tree model, including:

[0090] Based on the first lower-level event failure probability density function, according to the input rule of the multi-dimensional TS dynamic gate, a second lower-level event failure probability density function is obtained;

[0091] Obtaining a first subordinate event probability distribution function according to the second subordinate event failure probability density function;

[0092] Based on the first upper-level event failure probability density function, according to the output rule of the multi-dimensional TS dynamic gate, the second upper-level event failure probability density function is obtained;

[0093] The first superior event probability distribution function is obtained according to the second superior event failure probability density function.

[0094] Specifically, when the lower level event X i The execution possibility of TS dynamic gate input rule l under the influence of working time and k influencing factors (h1, h2, ..., h2) other than working time for:

[0095]

[0096] Middle: f i (t i , h1, h2, …, h k ) is the lower-level event X in the input rule l under the influence of multiple factors i The failure probability density function of

[0097] O (l) is the order rule in rule l.

[0098] When the influencing factors are independent of each other, the probability distribution function of the lower-level event (the first lower-level event failure probability distribution function) F i (t i , h1, h2, …, h k )for:

[0099]

[0100] Among them: F i (t i ) is the subordinate event X under the influence of working hours i The failure probability distribution function of

[0101] F i (hρ ) in h ρ Subordinate events under the influence of factor X i The failure probability distribution function

[0102] The failure probability distribution function F of the lower-level event i (t i , h1, h2, …, h k ) and its failure probability density function (failure probability density function of the second lower level event) f i (t i , h1, h2, …, h k ) is:

[0103]

[0104] Based on the input rule algorithm, the output rule algorithm of the TS dynamic gate under the influence of multiple factors is defined, and then the fault probability density function of the upper event y (the second upper event fault probability density function) f is calculated. y (t i , h1, h2, …, h k )for:

[0105]

[0106] Failure probability density function f for the superordinate event y y (t y , h1, h2, …, h k ) integration, and obtain the fault probability distribution function of the upper-level event y (the first upper-level event fault probability distribution function) F y (t y , h1, h2, …, h k )for:

[0107]

[0108] Further, the second fault probability distribution function is obtained based on the first probability distribution function and according to the third TS dynamic fault tree model, including:

[0109] Based on the first subordinate event probability distribution function, according to the input rule under the influence of polymorphism, a second subordinate event probability distribution function is obtained;

[0110] Based on the first upper-level event probability distribution function, according to the output rule under the influence of polymorphism, a second upper-level event probability distribution function is obtained;

[0111] According to the second lower-level event probability distribution function and the second upper-level event probability distribution function, a top-level event failure probability distribution function is obtained.

[0112] Specifically, given the multi-dimensional reliability data of the state changes and occurrence sequence of subordinate events, the execution possibility of the input rules can be obtained through the input rule algorithm.

[0113] In the input rule l, the subordinate event X i The status of becomes Reliability data is Available Its meaning is X i state The failure probability distribution function of (the second lower-level event failure probability distribution function).

[0114]

[0115] Represents X i The status of becomes The probability of all possible paths is obtained, and the total number of possible paths is strip; Represents X i The status of becomes In becoming The probability of all possible paths can be calculated as:

[0116]

[0117]

[0118] The numbers in the superscript represent the status.

[0119] The probability distribution function of multidimensional faults at a certain moment is obtained, and the execution possibility of input rule l for:

[0120]

[0121] Based on the execution possibility of the input rules obtained by the input rule algorithm and combined with the output rule algorithm, the multi-dimensional reliability data of the upper-level event, namely the fault probability density function and the fault probability distribution function, can be obtained.

[0122] The state of the parent event y is determined by becomes The fault probability density function (fault probability distribution function of the second-level event) is:

[0123]

[0124] Among them: from rule l, o(t1), o(t2),…, o(t n) determines the order and range of multiple integrals.

[0125] The state of the parent event y is determined by becomes The failure probability distribution function is obtained:

[0126]

[0127] The output rule reliability data of the lower TS dynamic gate is used to calculate the input rule reliability data of the upper T-·S dynamic gate. The reliability data of the top event T is finally obtained, that is, the T state is becomes The failure probability density function and failure probability distribution function of .

[0128] Furthermore, the importance of each event includes:

[0129] Probability importance, key importance, FV importance and comprehensive importance.

[0130] Specifically, the TS dynamic fault tree importance is obtained by multi-dimensional and multi-state TS dynamic fault tree importance, and the probability importance, critical importance, FV importance and comprehensive importance are obtained according to the continuous time algorithm.

[0131] like Figure 5 As shown in the figure, for the multi-dimensional multi-state TS dynamic fault tree, the probability importance and key importance are proposed to describe the impact of the basic event state on the top event state. It is a function of factors such as time, basic event reliability parameters and system structure. It can analyze the relative importance of each basic event at the same reliability level and evaluate the impact of the current basic event on the top event at different levels. The greater the importance of the basic event, the weaker the link of the basic event, the easier it is to cause system failure, and the more worthy of attention its potential risk is.

[0132] Specifically, the probability importance is used to describe the impact of a component changing from a normal working state to a faulty state on the system reliability. It is the difference in failure probability of the top event when the basic event is in a faulty state and when the basic event is in a normal state. It can reflect the status of the component in the system and is one of the bases for judging weak links.

[0133] The essence of probability importance is the probability distribution function of top event T to basic event x i The partial derivative of the failure probability distribution function is x i Status is For T status The probability importance of is:

[0134]

[0135] in: The T state is respectively becomes x i Status by becomes The failure probability distribution function.

[0136] Basic Event x i The status of the top event T is The probability importance of

[0137] Specifically, critical importance is used to measure the impact of changes in the probability of basic event failures on changes in the probability of top event failures. Compared with probability importance, it emphasizes the failure possibility of the component itself more and considers the relationship between component quality and system reliability. It is an important basis for system improvement and optimization.

[0138] The key importance introduces the basic event x on the basis of the probability importance. i The ratio of the failure probability distribution function of the top event T to the failure probability distribution function of the top event T, then x i Status is For T status The key importance is:

[0139]

[0140] Basic Event x i The status of the top event T is The key importance for

[0141]

[0142] Specifically, FV importance is the maximum value of the increase in the top event failure probability caused by each fault state of the basic event. It is used to evaluate the potential risks of system operation and find the components that cause the greatest hidden dangers of failure to the system. It is an important basis for system diagnosis and maintenance.

[0143] FV importance is based on basic event x i The possibility of failure is taken as a condition, and the difference of the failure probability distribution function of the top event T under the two conditions is calculated, then x i For T status The FV importance is:

[0144]

[0145] in: Represents x i Status is Under the condition of T state becomes The failure probability distribution function.

[0146] Specifically, the comprehensive importance takes into account the continuity of event state changes and introduces a superior state based on the probability importance. The basic event x i The probability of occurrence of a state and its instantaneous failure rate λ i The product of x i Status is The status of the top event T is The comprehensive importance of is:

[0147]

[0148] Basic Event x i The status of the top event T is The overall importance of We can get:

[0149]

[0150] Specifically, the first higher-level event failure probability density function expression is:

[0151]

[0152] Among them, f i (t i ) is the probability density function of the first lower-level event failure, f i (t y ) is the probability density function of the first upper-level event failure, i is the failure time of the lower-level event, t i is the failure time of the ith subordinate event, t y The upper-level event occurs at the failure time of the lower-level event, f i is the failure probability density function.

[0153] Specifically, the expression of the first subordinate event probability distribution function is:

[0154]

[0155] Where k is the number of image factors, h K represents the Kth influencing factor, F i (h ρ ) is h ρ Subordinate events under the influence of factor X i The failure probability distribution function.

[0156] Specifically, the expression of the probability distribution function of the first superior event is:

[0157]

[0158] Specifically, the expression of the second subordinate event probability distribution function is:

[0159]

[0160] in, The status of the subordinate event is represented by becomes The probability of all possible paths, S represents the state of the event; The status of the subordinate event is represented by becomes In becoming The probability of all possible paths, A is a certain state. A takes values ​​of (2, 3, ..., n).

[0161] Specifically, the expression of the second superior event probability distribution function is:

[0162]

[0163] in, The state of the parent event y is determined by becomes The failure probability density function, q is the qth influencing factor, s y is a state of the top event, b y The probability function of the fault state occurrence of the superior event.

[0164] The present invention provides a specific embodiment:

[0165] Step 1: Define the interrelationships of events at all levels of the wind turbine system

[0166] Step 2: Introduce continuous time description rules and establish a dynamic fault tree model

[0167] Step 3: Based on step 2, introduce the input and output rules of the multi-dimensional TS dynamic gate

[0168] Step 4: Based on step 3, introduce the input and output rules under the influence of polymorphism, and obtain the fault distribution function of the top event

[0169] Step 5: Obtain the importance of each basic event from the failure probability distribution function of the top event in step 4.

[0170] Specifically, the various operating states include shutdown, frequency reduction and normal; the various influencing factors include: working hours and other various influencing factors.

[0171] The beneficial effects of the present invention are as follows:

[0172] The present invention provides a reliability assessment method for a wind turbine generator set based on a TS fault tree, comprising: determining the mutual relationship between events at various levels of a wind turbine generator set system to be tested and the unit fault; based on the mutual relationship between events at various levels of the wind turbine generator set system to be tested and the unit fault, constructing a first TS dynamic fault tree model according to a unit step function; the events at various levels of the wind turbine generator set system to be tested include: lower-level events, upper-level events and top events; according to the first TS dynamic fault tree model, obtaining a first fault probability density function; the first fault probability density function includes: a first lower-level event fault probability density function and a first upper-level event fault probability density function; based on the first TS dynamic fault tree model, constructing a second TS dynamic fault tree model according to the input and output rules of a multidimensional TS dynamic gate; based on the first A fault probability density function, according to the second TS dynamic fault tree model, a first probability distribution function is obtained; based on the second TS dynamic fault tree model, the input and output rules under the influence of multiple states are introduced to construct a third TS dynamic fault tree model; based on the first probability distribution function, according to the third TS dynamic fault tree model, a second fault probability distribution function is obtained; the second fault probability distribution function is a top event fault probability distribution function; according to the second fault probability distribution function, the importance of each event is obtained; according to the importance, the wind turbine to be tested is evaluated. The present invention can characterize any dynamic failure behavior through TS dynamic gates and continuous time description rules, and an algorithm is given to solve the top event fault probability, which makes up for the shortcomings of the dynamic fault tree in terms of description means and calculation methods. In addition, the input and output rules of the TS dynamic fault tree under multiple external influencing factors and the three states of wind turbine shutdown, frequency reduction and normal operation are introduced, and thus the conversion probability distribution function of the wind turbine in each state is obtained, and the conversion probability distribution function of the wind turbine in each state under multiple influencing factors is obtained, which makes up for the diversification and multi-angle of reliability evaluation. Finally, the probability distribution function of wind turbine failure is used to obtain four reliability indices of probability importance, critical importance, FV importance and comprehensive importance to evaluate the impact of the part on the whole and to reflect the impact of each subsystem of the wind turbine on the wind turbine failure.

[0173] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0174] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A reliability assessment method for wind turbines based on TS fault tree, characterized in that: include: Determine the relationship between various events of the wind turbine system to be tested and the unit faults; The events include blade failure, hub failure, lightning protection device failure, pitch motor failure, pitch reducer failure, pitch bearing and pinion failure, rotary encoder failure, stator assembly failure, rotor assembly failure, main shaft failure, yaw motor and reducer failure, yaw bearing failure damage, yaw brake disc failure, yaw brake failure, natural aging and rust, measured wind direction is inconsistent with the actual wind direction, measured wind speed is inconsistent with the actual wind speed, repeated action of alternating loads and tower failure; Based on the relationship between the events at all levels of the wind turbine system to be tested and the unit faults, the first TS dynamic fault tree model is constructed according to the unit step function; The various levels of events of the wind turbine system to be tested include: lower level events, upper level events and top events; According to the first TS dynamic fault tree model, a first fault probability density function is obtained; the first fault probability density function includes: a first lower-level event fault probability density function and a first upper-level event fault probability density function; Based on the first TS dynamic fault tree model, a second TS dynamic fault tree model is constructed according to the input and output rules of the multi-dimensional TS dynamic gate; Based on the first fault probability density function, according to the second TS dynamic fault tree model, a first probability distribution function is obtained; Based on the second TS dynamic fault tree model, the input and output rules under the influence of polymorphism are introduced to construct a third TS dynamic fault tree model; Based on the first probability distribution function, according to the third TS dynamic fault tree model, a second fault probability distribution function is obtained; the second fault probability distribution function is a top event fault probability distribution function; According to the second fault probability distribution function, the importance of each event is obtained; Evaluate the wind turbine to be tested according to the importance; The first probability distribution function comprises: A first lower-level event probability distribution function and a first upper-level event probability distribution function; The method of obtaining a first probability distribution function based on the first fault probability density function and according to the second TS dynamic fault tree model comprises: Based on the first lower-level event failure probability density function, according to the input rule of the multi-dimensional TS dynamic gate, a second lower-level event failure probability density function is obtained; Obtaining a first subordinate event probability distribution function according to the second subordinate event failure probability density function; Based on the first upper-level event failure probability density function, according to the output rule of the multi-dimensional TS dynamic gate, the second upper-level event failure probability density function is obtained; Obtaining a first superior event probability distribution function according to the second superior event failure probability density function; The method of obtaining a second fault probability distribution function based on the first probability distribution function and according to the third TS dynamic fault tree model includes: Based on the first subordinate event probability distribution function, according to the input rule under the influence of polymorphism, a second subordinate event probability distribution function is obtained; Based on the first upper-level event probability distribution function, according to the output rule under the influence of polymorphism, a second upper-level event probability distribution function is obtained; According to the second lower-level event probability distribution function and the second upper-level event probability distribution function, a top-level event failure probability distribution function is obtained.

2. A reliability assessment method for wind turbines based on TS fault tree according to claim 1, characterized in that: The importance of each event includes: Probability importance, key importance, FV importance and comprehensive importance.

3. A reliability assessment method for wind turbines based on TS fault tree according to claim 1, characterized in that: The first higher-level event failure probability density function expression is: Among them, f i (t i ) is the probability density function of the first lower-level event failure, f i (t y ) is the probability density function of the first upper level event failure, t i is the failure time of the ith subordinate event, t y is the failure time of the superior event, f i is the failure probability density function, δ(t i -t y ) is the unit pulse function.

4. A reliability assessment method for wind turbines based on TS fault tree according to claim 3, characterized in that: The expression of the probability distribution function of the first subordinate event is: Among them, k is the number of influencing factors, h k represents the kth influencing factor, F i (h ρ ) is h ρ Subordinate events under the influence of factor X i The failure probability distribution function.

5. A reliability assessment method for wind turbines based on TS fault tree according to claim 4, characterized in that: The expression of the probability distribution function of the first superior event is:

6. A reliability assessment method for wind turbines based on TS fault tree according to claim 5, characterized in that: The expression of the second subordinate event probability distribution function is: ； in, The status of the subordinate event is represented by becomes The probability of all possible paths, S represents the state of the event; The status of the subordinate event is represented by becomes Then become The probability of all possible paths, g = 1, 2..., k i -A; A is a state.

7. A reliability assessment method for wind turbines based on TS fault tree according to claim 6, characterized in that: The expression of the second superior event probability distribution function is: ； Among them, h q is the qth influencing factor, s y is a state of the top event, b y is the probability function of the fault state occurrence of the superior event.

Images