A floating wind turbine related component failure risk analysis method and system

Abstract

The application discloses a floating wind turbine related component failure risk analysis method and system, and the method comprises the following steps: step S1, a Bayesian network model of key components of an offshore floating wind turbine under common cause failure is established, and a conditional probability table of a sub-node thereof is calculated; step S2, Bayesian inference is adopted to respectively obtain an initial state of a Bayesian network node and a failure probability of a key component after evidence updating; step S3, a failure correlation degree of each key component of the wind turbine is calculated, and a failure risk priority number based on an expert scoring method is updated, and finally, a list of key components of the offshore floating wind turbine with high failure risks that need to be focused on is screened and determined. The reliability analysis credibility in the application will gradually improve with the continuous expansion of failure statistical prior data, and the application can be extended to different stages of the whole life cycle of the offshore floating wind turbine to improve the operation reliability thereof.

Description

TECHNICAL FIELD

[0001] The present application relates to a system reliability calculation method for offshore floating wind turbines under common cause failure, in particular to a floating wind turbine related component failure risk analysis method and system. BACKGROUND

[0002] Offshore floating wind turbines are newly emerging wind energy capture conversion equipment. Because of the ability to operate in deeper and more distant sea areas, offshore floating wind turbines are considered by the industry as one of the promising offshore wind energy development equipment, and have a certain irreplaceability in deep-sea wind energy exploration. The development potential of offshore floating wind turbines depends heavily on their high reliability, high availability and lower failure rate, i.e. producing more electricity safely and with a lower failure rate in the entire life cycle.

[0003] Reliability analysis of offshore floating wind turbines is the basis for identifying offshore floating wind turbine failure characteristics, failure modes, failure causes, and failure avoidance measures, and will directly support the realization of high reliability, high availability, and lower failure rate of offshore floating wind turbines. The current existing reliability analysis methods suitable for offshore floating wind turbines are generally based on the assumption of independence of components, i.e. the components of offshore floating wind turbines are independent of each other and there is no mutual influence between the failures of the components. However, offshore floating wind turbines are complex systems composed of mechanical components, electrical components, electronic components, hydraulic components, sensors, etc. The failure forms are complex and have correlations, and are mostly caused by common cause failures. For example, sea wind and sea wave are the causes of blade, floating platform, and mooring system failures, so the failures of blades, floating platforms, and mooring systems have correlations caused by common cause failures; the central lubrication system simultaneously provides lubrication for the pitch system, yaw system, and gearbox, so the failures of the pitch system, yaw system, and gearbox have correlations caused by common cause failures. SUMMARY

[0004] The purpose of the present application is to overcome the above-mentioned problems in the prior art, and to provide a floating wind turbine related component failure risk analysis method and system, which can well model the correlation between component failures into the reliability analysis of key components of offshore floating wind turbines.

[0005] In order to achieve the above-mentioned purpose, the present application is realized by the following technical scheme:

[0006] A floating wind turbine related component failure risk analysis method, comprising the following steps:

[0007] Step S1, establishing a Bayesian network model of key components of offshore floating wind turbines under common cause failure and calculating the conditional probability table of the child nodes;

[0008] Step S2, the initial state of the Bayesian network node and the failure probability of the key components after the evidence update are obtained by Bayesian inference respectively;

[0009] Step S3, the failure correlation degree of each key component of the wind turbine is calculated, and the failure risk priority number based on the expert scoring method is updated, and finally the list of key components of the offshore floating wind turbine with high failure risk is screened and determined.

[0010] As preferred, the key components of the offshore floating wind turbine are k key components, including gearbox, generator, floating foundation, mooring, wherein n key components exist common cause failure, and the Bayesian network model of the key components of the offshore floating wind turbine under common cause failure is:

[0011] B=<G,A CPT >

[0012] Wherein, G represents the network structure of the Bayesian network model of the key components of the offshore floating wind turbine; A CPT represents the conditional probability table set of the Bayesian network model; the expression of G is:

[0013] G=<V,E>

[0014] Wherein, V={r,c1,c2,L,c j ,L,c n (n≤k)} represents the set of nodes in the network structure G, r represents a root node, representing the state of the common cause failure component; c1,c2,L,c j ,L,c n represent n child nodes with the same failure cause; E is the set of directed edges in the network structure G; the root node r in the Bayesian network model is connected to the n child nodes through n directed edges, and the expression of r in c1,c2,L,c j ,L,c n is:

[0015]

[0016] Wherein, Fault(r=0) and Normal(r=1) respectively represent the failure and normal operation of the common cause failure component of the offshore floating wind turbine.

[0017] The expression of c j in c1,c2,L,c j ,L,c n is:

[0018]

[0019] Wherein, Fault(c j =0) and Normal(cj = 1) represent the jth key component failure and normal operation of offshore floating wind turbine, respectively;

[0020] The A CPT is:

[0021] A CPT = {a1, a2, L, a j , L, a n}

[0022] Wherein, a j is the conditional probability table of the jth sub-node, n is the number of key components in the Bayesian network model of offshore floating wind turbine key components; a j The expression is:

[0023]

[0024] Wherein, the definition of four conditional probabilities is:

[0025] P(c j = 0 | r = 0) is the conditional probability of the jth key component failure corresponding to the sub-node when the common cause failure component fails and the corresponding root node fails in the Bayesian network model; P(c j = 0 | r = 1) is the conditional probability of the jth key component failure corresponding to the sub-node when the common cause failure component fails and the corresponding root node is normal in the Bayesian network model; P(c j = 1 | r = 0) is the conditional probability of the jth key component normal operation corresponding to the sub-node when the common cause failure component fails and the corresponding root node fails in the Bayesian network model; P(c j = 1 | r = 1) is the conditional probability of the jth key component normal operation corresponding to the sub-node when the common cause failure component fails and the corresponding root node is normal in the Bayesian network model.

[0026] As preferred, the four conditional probability calculation steps are:

[0027] (1) From the offshore floating wind turbine fault work order, obtain the total failure times N, the monitoring time T, the failure frequency f r of the common cause failure component r, the failure frequency of the jth key component, the frequency of simultaneous failure of the common cause failure component r and the jth key component

[0028] (2) Calculate the failure probability P(r = 0) of the common cause failure component:

[0029]

[0030] (3) Calculate the normal operation probability P(r = 1) of the common cause failure component:

[0031]

[0032] (4) Calculate the probability P(c j =0 & r=0) that the jth common cause failure component and the key component fail simultaneously:

[0033]

[0034] (5) Calculate the probability P(c j =0 & r=1) that the common cause failure component operates normally and the jth key component fails simultaneously:

[0035]

[0036] (6) The four conditional probability expressions in a j are:

[0037]

[0038]

[0039]

[0040]

[0041] As a preferred, the Bayesian inference is used in step S2 to obtain the initial state of the Bayesian network node and the failure probability of the key component after the evidence is updated, respectively, and the steps further include:

[0042] S21, calculate the failure probability P(c j =0) of the jth key component in the initial state:

[0043] P(c j =0) = P(c j =0 | r=0) · P(r=0) + P(c j =0 | r=1) · P(r=1);

[0044] S22, calculate the normal operation probability P(c j =1) of the jth key component in the initial state:

[0045] P(c j =1) = 1 - P(c j =0);

[0046] S23, after the state of the jth key component in the Bayesian network model nodes c1, c2, L, c j , L, c n is updated, the failure probability P update(c j = 1) = 1, the probability of normal operation P update (c j = 0) = 0;

[0047] S24, calculate the failure probability P j of the common cause failure component after updating the state evidence of the jth key component in c1, c2, L, c n update (r = 0):

[0048]

[0049] S25, calculate the probability P j of normal operation of the common cause failure component after updating the state evidence of the jth key component in c1, c2, L, c n update (r = 1):

[0050] P update (r = 1) = 1 - P update (r = 0);

[0051] S26, calculate the failure probability P j of the ith (i ≠ j) key component after updating the state evidence of the jth key component in c1, c2, L, c n update (c i = 0):

[0052] P update (c i = 0) = P(c i = 0 | r = 0) · P update (r = 0) + P(c i = 0 | r = 1) · P update (r = 1);

[0053] As a preferred, in step S3, the failure correlation degree of each key component of the wind turbine is calculated, and the failure risk priority number based on the expert scoring method is updated, and finally the list of key components of the offshore floating wind turbine with high failure risk which is focused on is screened and determined, further comprising the following steps:

[0054] S31, calculate the failure correlation degree △P j of the jth key component:

[0055]

[0056] Wherein, the failure correlation degree of the jth key component is represented as c1, c2, L, c j n ​​​​not including c j the sum of the variation of the failure probability of the n-1 key components.

[0057] S32, calculating the failure risk priority number RPN of the key components of the offshore floating wind turbine based on the expert scoring method j :

[0058]

[0059] wherein RPN j is the failure risk priority number of the jth key component of the offshore floating wind turbine; is the score given by the domain expert for the ith risk evaluation criterion of the jth key component of the offshore floating wind turbine; is the weight given by the domain expert for the ith failure attribute of the jth key component of the offshore floating wind turbine; respectively the severity, occurrence and detectability of the failure risk in the jth key component of the offshore floating wind turbine; k is the total number of key components of the offshore floating wind turbine.

[0060] S33, updating the failure risk priority number of the jth key component of the offshore floating wind turbine:

[0061]

[0062] According to the value of RPN update,j the k key components of the offshore floating wind turbine are arranged in descending order according to the value size, thereby obtaining a list of high-fault-risk components of the offshore floating wind turbine considering the related failure.

[0063] As a preferred, the system comprises:

[0064] a building module for building a Bayesian network model of the key components of the offshore floating wind turbine under common cause failure;

[0065] a probability calculation module for calculating the conditional probability table of the child nodes;

[0066] an acquisition module for acquiring the failure probability of the key components after the initial state and the evidence are updated;

[0067] a failure calculation module for calculating the failure correlation degree of each key component of the wind turbine;

[0068] an updating module for updating the failure risk priority number based on the expert scoring method;

[0069] The screening module is used for screening and determining a list of key components of the offshore floating wind turbine with high failure risks. The offshore floating wind turbine is analyzed in reliability, and the reliability analysis credibility of the offshore floating wind turbine is gradually improved with the continuous expansion of the failure statistical prior data, and the offshore floating wind turbine can be expanded to different stages of the whole life cycle to improve the operation reliability. BRIEF DESCRIPTION OF DRAWINGS

[0070] Figure 1 The flowchart of the present application;

[0071] Figure 2 The Bayesian network model diagram of the key components of the offshore floating wind turbine under related failure. DETAILED DESCRIPTION

[0072] The present application will be further described in detail below with reference to the accompanying drawings and embodiments.

[0073] As shown in the following steps: Figure 1

[0074] Step S1, a Bayesian network model of the key components of the offshore floating wind turbine under common cause failure is established, and a conditional probability table of a child node is calculated;

[0075] Step S2, the initial state of the Bayesian network node and the failure probability of the key components after the evidence is updated are obtained by using Bayesian inference;

[0076] Step S3, the failure correlation degree of each key component of the wind turbine is calculated, the failure risk priority number based on the expert scoring method is updated, and finally the list of key components of the offshore floating wind turbine with high failure risks is screened and determined.

[0077] Further, the step S1 of establishing the Bayesian network model of the key components of the offshore floating wind turbine under common cause failure and calculating the conditional probability table of the child node further comprises:

[0078] S11, the offshore floating wind turbine has k key components including a gearbox, a generator, a floating foundation, a mooring, etc., and n key components exist under common cause failure, and the Bayesian network model of the key components of the offshore floating wind turbine under common cause failure is:

[0079] B = < G, A CPT > ​

[0080] Where G represents the network structure of the Bayesian network model of the key components of the offshore floating wind turbine; A CPT G represents the set of conditional probability tables for a Bayesian network model; its expression is:

[0081] G =<V,E>

[0082] Where V={r,c1,c2,L,c j ,L,c n (n≤k)} represents the set of nodes in the network structure G, r represents a root node, and represents the state of the common-cause failure component; c1, c2, ..., c j ,L,c n Let c1, c2, ..., cn be the n child nodes that share the same failure cause; let E be the set of directed edges in the network structure G; in this Bayesian network model, the root node r is connected to each of the n child nodes c1, c2, ..., cn by n directed edges. j ,L,c n The expression for r is:

[0083]

[0084] Here, Fault (r=0) and Normal (r=1) represent the failure of common-cause components and normal operation of offshore floating wind turbines, respectively.

[0085] Nodes c1, c2, L, c j ,L,c n c j The expression is:

[0086]

[0087] Among them, Fault(c j =0) and Normal(c j =1) represent the failure and normal operation of the key components of the j-th offshore floating wind turbine, respectively.

[0088] S12, The set of conditional probability tables for Bayesian network models of key components of offshore floating wind turbines (A) CPT )for:

[0089] A CPT ={a1,a2,L,a j ,L,a n}

[0090] Among them, a j Let a be the conditional probability table for the j-th child node, and n be the number of key components in the Bayesian network model of key components of offshore floating wind turbines; a j The expression is:

[0091]

[0092] wherein the four conditional probabilities are defined as:

[0093] P(c j =0|r=0) is the conditional probability of the failure of the jth key component when the common cause failure component fails and the corresponding root node fails in the Bayesian network model; P(c j =0|r=1) is the conditional probability of the failure of the jth key component when the common cause failure component fails and the corresponding root node is in normal operation in the Bayesian network model; P(c j =1|r=0) is the conditional probability of the normal operation of the jth key component when the common cause failure component fails and the corresponding root node fails in the Bayesian network model; P(c j =1|r=1) is the conditional probability of the normal operation of the jth key component when the common cause failure component fails and the corresponding root node is in normal operation in the Bayesian network model.

[0094] The calculation steps of the four conditional probabilities are as follows:

[0095] (1) Obtain the total failure number N, the monitoring time T, the failure frequency f r of the common cause failure component r, and the failure frequency f of the jth key component from the offshore floating wind turbine failure work order.

[0096] (2) Calculate the failure probability P(r=0) of the common cause failure component:

[0097]

[0098] (3) Calculate the normal operation probability P(r=1) of the common cause failure component:

[0099]

[0100] (4) Calculate the probability P(c j =0&r=0) that the jth common cause failure component and the key component fail simultaneously:

[0101]

[0102] (5) Calculate the probability P(c j =0&r=1) that the common cause failure component is in normal operation and the jth key component fails simultaneously:

[0103]

[0104] (6)j The four conditional probability expressions in the Bayesian network model are as follows:

[0105]

[0106]

[0107]

[0108]

[0109] In the embodiment of the present application, the key components of the offshore floating wind turbine are shown in Table 1, the Bayesian network model of the key components of the offshore floating wind turbine is shown in Table 2, and the conditional probability of each sub-node of the Bayesian network model of the key components of the offshore floating wind turbine is shown in Table 2. Figure 2

[0110] Table 1: List of key components of offshore floating wind turbine

[0111]

[0112] Table 2: Conditional probability table of sub-nodes of Bayesian network model of key components of offshore floating wind turbine

[0113]

[0114] Further, the step S2 of obtaining the failure probability of the key components in the initial state and after the evidence update by using Bayesian inference further comprises:

[0115] S21, calculating the failure probability P(c j =0) of the jth key component in the initial state:

[0116] P(c j =0) = P(c j =0 | r = 0) · P(r = 0) + P(c j =0 | r = 1) · P(r = 1)

[0117] S22, calculating the normal operation probability P(c j =1) of the jth key component in the initial state:

[0118] P(c j =1) = 1 - P(c j =0)

[0119] S23, after updating the state of the jth key component in the Bayesian network model nodes c1, c2, L, c j , L, c n , the failure probability P update (c j ​= 1, the probability of normal operation P update (c j = 0) = 0;

[0120] S24, calculate the failure probability P j ,L, c n of the ith key component after the state evidence of the jth key component in c1, c2, L, c update (r = 0) is updated:

[0121]

[0122] S25, calculate the probability P j ,L, c n of normal operation of the ith key component after the state evidence of the jth key component in c1, c2, L, c update (r = 1) is updated:

[0123] P update (r = 1) = 1 - P update (r = 0)

[0124] S26, calculate the failure probability P j ,L, c n of the ith (i ≠ j) key component after the state evidence of the jth key component in c1, c2, L, c update (c i = 0) is updated:

[0125] P update (c i = 0) = P(c i = 0 | r = 0) · P update (r = 0) + P(c i = 0 | r = 1) · P update (r = 1)

[0126] In the embodiment of the application, the conditional probability table of the key component of the offshore floating wind turbine obtained in step S1 is substituted into the Bayesian inference of step S2 to obtain the initial state of the Bayesian network of the key component of the offshore floating wind turbine. The conditional probability table result of step S1 is used to perform Bayesian inference to obtain the failure probability of the key component after evidence update. The initial state of the Bayesian network of the key component of the offshore floating wind turbine and the failure probability of the key component after evidence update are shown in Table 3.

[0127] Table 3 Initial state of Bayesian network of key component of offshore floating wind turbine and failure probability of key component after evidence update

[0128]

[0129] Further, the failure correlation degree of each key component of the wind turbine is calculated in step S3, and the failure risk priority number based on the expert scoring method is updated, and finally the list of key components of the offshore floating wind turbine with high failure risk that needs to be focused on is screened and determined, further comprising the following steps:

[0130] S31, calculate the failure correlation degree △P of the jth key component j :

[0131]

[0132] Wherein, the failure correlation degree of the jth key component is expressed as c1, c2, L, c j , L, c n The sum of the failure probability change of the other n-1 key components excluding c j in the jth key component;

[0133] S32, calculate the failure risk priority number RPN of the key components of the offshore floating wind turbine based on the expert scoring method j :

[0134]

[0135] Wherein, RPN j is the failure risk priority number of the jth key component of the offshore floating wind turbine; is the score given by the field expert for the ith risk evaluation criterion of the jth key component of the offshore floating wind turbine; is the weight given by the field expert for the ith failure attribute of the jth key component of the offshore floating wind turbine; The severity, occurrence and detection of the failure risk in the failure attribute of the jth key component of the offshore floating wind turbine; k is the total number of key components of the offshore floating wind turbine;

[0136] S33, update the failure risk priority number of the jth key component of the offshore floating wind turbine:

[0137]

[0138] According to the value of RPN update,j , the k key components of the offshore floating wind turbine are arranged in descending order, and the list of high failure risk components of the offshore floating wind turbine considering the related failure can be obtained.

[0139] In the embodiment of the present application, the initial state of the Bayesian network of the key components of the offshore floating wind turbine in step S2 and the updated failure probability of the key components are converted into the failure correlation degree of the key components of the offshore floating wind turbine by step S31, as shown in Table 4. The calculation result of the offshore floating wind turbine key component failure risk priority number index based on the expert scoring method is obtained by step S32, as shown in Table 5. The failure correlation degree of the offshore floating wind turbine key components obtained by S31 and the risk priority number ranking result of the offshore floating wind turbine key components based on the expert scoring method obtained by S32 are substituted into step S3 to calculate the new risk priority number ranking of the offshore floating wind turbine key components considering the related failure, as shown in Table 6. The list of offshore floating wind turbine components with high failure risk is obtained by arranging the updated risk priority number in descending order, as shown in Table 7.

[0140] Table 4 Failure correlation degree of offshore floating wind turbine key components

[0141]

[0142] Table 5 RPN ranking of offshore floating wind turbine key components based on expert scoring method

[0143]

[0144]

[0145] Table 6 Failure risk priority number ranking table of offshore floating wind turbine key components considering failure correlation

[0146]

[0147] Table 7 List of offshore floating wind turbine components with high failure risk

[0148]

[0149] The reliability analysis method of the offshore floating wind turbine considering the related failure of the present application is used to determine the failure risk level and its ranking at the key component level. In general, the present application can more accurately and effectively carry out reliability analysis of offshore floating wind turbines and obtain more realistic, reliable and accurate reliability analysis results.

[0150] It should be noted that the above is only one specific embodiment of the present application. Obviously, the present application is not limited to the above embodiment, but can have many variations. In general, all variations that can be directly derived or inferred from the disclosed content by those skilled in the art should be considered as falling within the scope of the present application.

Claims

1. A method for failure risk analysis of relevant components of a floating wind turbine, characterized in that, Includes the following steps: Step S1: Establish a Bayesian network model of key components of offshore floating wind turbines under common cause failure and calculate its child node conditional probability table; Step S2: Use Bayesian inference to obtain the initial state of the Bayesian network nodes and the failure probability of the key components after the evidence update. Step S3: Calculate the failure correlation of each key component of the wind turbine, update the failure risk priority number based on the expert scoring method, and finally screen and determine the list of key components of offshore floating wind turbines with high failure risk. The offshore floating wind turbine has k key components, including a gearbox, generator, floating foundation, and mooring. N of these key components experience common cause failure. Therefore, the Bayesian network model of the key components of the offshore floating wind turbine under common cause failure is: B=<G,A CPT >; Where G represents the network structure of the Bayesian network model of the key components of the offshore floating wind turbine; A CPT G represents the set of conditional probability tables for a Bayesian network model; its expression is: G =<V,E> Where V = {r, c1, c2, ..., c} j c n (n≤k)} represents the set of nodes in the network structure G, r represents a root node, and represents the state of the common cause failure component; c1, c2, ..., c j c n Let represent n child nodes with the same failure cause; E is the set of directed edges in the network structure G; the root node r in this Bayesian network model is connected to n child nodes c1, c2, ..., c3 by n directed edges. j c n The expression for r is: Where Fault (r=0) and Normal (r=1) represent the failure of common-cause components and normal operation of offshore floating wind turbines, respectively; c1, c2, ..., c j c n c j The expression is: Among them, Fault(c j =0) and Normal(c j =1) represent the failure and normal operation of the key components of the j-th offshore floating wind turbine, respectively; The set A of conditional probability tables for a Bayesian network model CPT for: A CPT ={a1,a2,…,a j ,…,a n }; Among them, a j Let a be the conditional probability table for the j-th child node, and n be the number of key components in the Bayesian network model of key components of offshore floating wind turbines; a j The expression is: The four conditional probabilities are defined as follows: P(c j =0|r=0) is the conditional probability of the critical component corresponding to the j-th child node failing when the corresponding root node fails due to the failure of a failed component in the Bayesian network model; P(c j =0|r=1) is the conditional probability of the critical component corresponding to the j-th child node failing when the corresponding root node is operating normally in the Bayesian network model due to the failure of a failed component; P(c j =1|r=0) is the conditional probability of the critical component corresponding to the j-th child node operating normally when the root node corresponding to the component failure fails due to failure in the Bayesian network model; P(c j =1|r=1) is the conditional probability of the critical component corresponding to the j-th child node operating normally when the corresponding root node is operating normally due to the failure of the failed component in the Bayesian network model.

2. The method for analyzing the failure risk of related components of a floating wind turbine according to claim 1, characterized in that, The calculation steps for the four conditional probabilities are as follows: (1) Obtain the total number of failures N, monitoring duration T, and failure frequency f of the common cause failure component r from the offshore floating wind turbine failure work orders. r The failure frequency of the j-th critical component The frequency of simultaneous failure of common-cause component r and the j-th critical component (2) Calculate the failure probability P(r=0) of the component with common cause failure: (3) Calculate the normal operating probability P(r=1) of the component that failed due to common cause: (4) Calculate the probability P(c) of the j-th common cause failure component and the critical component failing simultaneously. j =0&r=0): (5) Calculate the probability P(c) of the simultaneous occurrence of normal operation of the common-cause failure component and the failure of the j-th critical component. j =0&r=1): (6)a j The four conditional probability expressions in the text are:

3. The method for analyzing the failure risk of related components of a floating wind turbine according to claim 1, characterized in that, In step S2, Bayesian inference is used to obtain the initial state of the Bayesian network nodes and the failure probability of key components after evidence update. This step further includes: S21. Calculate the failure probability P(c) of the j-th critical component in the initial state. j =0): P(c j =0)=P(c j =0|r=0)·P(r=0)+P(c j =0|r=1)·P(r=1) S22. Calculate the normal operation probability P(c) of the j-th critical component in the initial state. j =1): P(c j =1)=1-P(c j =0) S23, Bayesian network model nodes c1, c2, ..., c j c n After the state of the j-th critical component is updated, the failure probability P of that component is... update (c j =0)=1, the probability P of normal operation update (c j =1)=0; S24. Calculate the condition when c1, c2, ..., c j c n After the status evidence of the j-th critical component is updated, the failure probability P of the component that failed due to common causes is... update (r=0): S25. Calculate the value of c1, c2, ..., c when c1, c2, ..., c j c n After the status evidence of the j-th critical component is updated, the probability P of the component that failed due to common cause operating normally is... update (r=1): P update (r=1)=1-P update (r=0) S26. Calculate the condition when c1, c2, ..., c j c n After the state evidence of the j-th critical component is updated, the failure probability P of the i-th (i≠j) critical component is... update (c i =0): P update (c i =0)=P(c i =0|r=0)·P update (r=0)+P(c i =0|r=1)· P update (r＝1)。 4. The method for analyzing the failure risk of related components of a floating wind turbine according to claim 1, characterized in that, Step S3 involves calculating the failure correlation degree of each critical component of the wind turbine, updating the failure risk priority number based on expert scoring, and finally screening and determining a list of critical components of offshore floating wind turbines with high failure risk. This further includes the following steps: S31. Calculate the failure correlation degree ΔP of the j-th critical component. j : Wherein, the failure correlation degree of the j-th critical component is represented as c1, c2, ..., c j c n The middle does not include c j The sum of the changes in the failure probability of the other n-1 key components, including the main component; S32. Calculate the failure risk priority number (RPN) for critical components of offshore floating wind turbines based on expert scoring method. j : Among them, RPN j The failure risk priority number for the j-th critical component of an offshore floating wind turbine; The score given by the domain experts for the i-th risk assessment criterion of the j-th critical component of the offshore floating wind turbine; The weight of the i-th failure attribute of the j-th critical component of an offshore floating wind turbine, as given by domain experts; These represent the severity, occurrence, and detectability of the failure risk in the failure attributes of the j-th critical component of the offshore floating wind turbine; k represents the total number of critical components of the offshore floating wind turbine. S33. Update the failure risk priority number of the j-th critical component of the offshore floating wind turbine: According to RPN update，j Arranging the k key components of an offshore floating wind turbine in descending order of numerical values ​​yields a list of high-failure-risk components for the offshore floating wind turbine, taking into account related failures.

5. A system based on the failure risk analysis method for relevant components of a floating wind turbine as claimed in claim 1, characterized in that, The system includes: Establishment Module: Used to establish Bayesian network models of key components of offshore floating wind turbines under common cause failure; The probability calculation module is used to calculate the conditional probability table of child nodes. Acquisition module: used to acquire the failure probability of key components in the initial state and after evidence updates; Failure Calculation Module: Used to calculate the failure correlation of each critical component of the wind turbine; Update module: Used to update the failure risk priority number based on expert scoring method; Screening module: Used to screen and identify a list of critical components of offshore floating wind turbines that are of high risk of failure; The offshore floating wind turbine has k key components, including a gearbox, generator, floating foundation, and mooring. N of these key components experience common cause failure. Therefore, the Bayesian network model of the key components of the offshore floating wind turbine under common cause failure is: B=<G,A CPT >; Where G represents the network structure of the Bayesian network model of the key components of the offshore floating wind turbine; A CPT G represents the set of conditional probability tables for a Bayesian network model; its expression is: G = <V,E> Where V = {r, c1, c2, ..., c} j c n (n≤k)} represents the set of nodes in the network structure G, r represents a root node, and represents the state of the common cause failure component; c1, c2, ..., c j c n Let represent n child nodes with the same failure cause; E is the set of directed edges in the network structure G; the root node r in this Bayesian network model is connected to n child nodes c1, c2, ..., c3 by n directed edges. j c n The expression for r is: Where Fault (r=0) and Normal (r=1) represent the failure of common-cause components and normal operation of offshore floating wind turbines, respectively; c1, c2, ..., c j c n c j The expression is: Among them, Fault(c j =0) and Normal(c j =1) represent the failure and normal operation of the key components of the j-th offshore floating wind turbine, respectively; The set A of conditional probability tables for a Bayesian network model CPT for: A CPT ={a1,a2,…,a j ,…,a n }; Among them, a j Let a be the conditional probability table for the j-th child node, and n be the number of key components in the Bayesian network model of key components of offshore floating wind turbines; a j The expression is: The four conditional probabilities are defined as follows: P(c j =0|r=0) is the conditional probability of the critical component corresponding to the j-th child node failing when the corresponding root node fails due to the failure of a failed component in the Bayesian network model; P(c j =0|r=1) is the conditional probability of the critical component corresponding to the j-th child node failing when the corresponding root node is operating normally in the Bayesian network model due to the failure of a failed component; P(c j =1|r=0) is the conditional probability of the critical component corresponding to the j-th child node operating normally when the root node corresponding to the component failure fails due to failure in the Bayesian network model; P(c j =1|r=1) is the conditional probability of the critical component corresponding to the j-th child node operating normally when the corresponding root node is operating normally due to the failure of the failed component in the Bayesian network model.

Images