Methods and terminals for quantifying the value of structural health monitoring information for helicopter rotor blades

Abstract

This invention discloses a method and terminal for quantifying the value of structural health monitoring information for helicopter rotor blades. The steps are as follows: establishing a rotor blade damage probability model based on Weibull failure distribution and Paris crack propagation formula; obtaining maintenance costs based on a damage cost model; and performing minimum cost decision analysis using a decision tree based on Bayesian update theory. This invention quantifies the value of deteriorated structural monitoring information by modeling structural performance and utility using a damage probability model, a Bayesian-based decision tree method, and a system cost and benefit analysis model. This solves the problem of difficulty in quantifying the economics of structural health monitoring, and is of great significance for operators to assess the impact on helicopter operations before investment and to promote the application of structural health monitoring.

Description

Technical Field

[0001] This invention belongs to the field of helicopter structural health monitoring technology, specifically relating to a method and terminal for quantifying the value of structural health monitoring information for helicopter rotor blades. Background Technology

[0002] To achieve preventative maintenance of mechanical systems, maintenance personnel can implement scheduled maintenance or condition-based maintenance. Traditional scheduled maintenance, in order to ensure a high level of system safety, often introduces safety margin redundancy, leading to excessively high maintenance costs. Compared to scheduled maintenance, condition-based maintenance develops maintenance plans based on the actual health status information of relevant components, offering greater adaptability and accuracy. Furthermore, helicopter rotor systems are complex, and the internal components cannot be visually inspected in a short time; this information can be accessed through structural health monitoring systems.

[0003] As a supplement to traditional periodic maintenance, fault prediction and health management utilizes sensor information, expert knowledge, and maintenance support information, along with various intelligent algorithms and models, to monitor and predict operational status. It has applications in multiple fields, such as ships and spacecraft. In the helicopter field, its application is a health and operational monitoring system, a complex system integrating avionics, ground support equipment, and onboard computer monitoring. It processes data such as vibration, azimuth, and flight parameters, outputting the status parameters of the monitored components to reflect their health status. The structural health and operational monitoring portion of HUMS has entered its fourth generation of operational development internationally since the 1990s. It incorporates a damage monitoring system on top of existing flight parameters to monitor local overloads and damage; this is the structural health monitoring system. Structural health monitoring is the process of identifying and characterizing structural damage. Its role in civil and aerospace engineering is recognized; however, related research in the domestic helicopter field has largely focused on damage detection, location, and feature extraction, with limited engineering applications. Compared to fixed-wing aircraft, helicopters have more complex loads, operate in harsher environments, and have a significantly higher accident rate. Furthermore, their maintenance and inspection costs are high, accounting for approximately 25% of the total operating cost. Structural health monitoring (SHealth) can monitor the structural health status in real time, including damage, loads, and deformation. This monitoring information facilitates condition-based maintenance by maintenance personnel. As a preventative auxiliary maintenance method, it can improve safety while reducing maintenance costs. Therefore, it is even more necessary for helicopters to adopt SHealth technology to take appropriate measures to mitigate damage propagation through early warning systems.

[0004] Rotor blades are subjected to significant dynamic loads and deflection moments under various operating conditions, and their deformation is a major source of lift for helicopters. Monitoring the vibration frequency at the blade root can provide more accurate condition information for maintenance, improving safety. It can also replace some routine inspections, extend maintenance intervals, and improve economic efficiency. Although structural health monitoring is widely recognized in the industry as a supplementary maintenance method, its application in helicopters is rare. One reason is that structural health monitoring systems are very expensive, and the information they provide is uncertain. Their economic benefits cannot be measured before actual deployment, and without clear economic benefits, it is often difficult to persuade decision-makers to invest. Therefore, quantifying the benefits of this monitoring system is crucial for promoting its application in the helicopter field, ensuring helicopter safety, improving maintenance efficiency, and guaranteeing availability. This is of great significance for advancing the development of helicopter structural monitoring. Summary of the Invention

[0005] To address the shortcomings of the existing technologies, the present invention aims to provide a method and terminal for quantifying the value of structural health monitoring information for helicopter rotor blades, thereby overcoming the problem of unclear economic benefits in existing structural health monitoring systems. By quantifying the economic benefits of monitoring information, this invention can accelerate its application and promotion in fields such as helicopters.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] The present invention provides a method for quantifying the value of structural health monitoring information for helicopter rotor blades, comprising the following steps:

[0008] 1) Establish a blade damage probability model based on Weibull failure distribution and Paris crack propagation formula: classify helicopter blade crack conditions into healthy state D1, repairable state D2, and unrepairable state D3; for the three crack conditions, there are three corresponding maintenance measures: no maintenance a1, repair crack a2, and blade replacement a3; according to the blade Weibull failure distribution and Paris crack propagation formula, perform probability modeling for the three crack conditions respectively.

[0009] 2) Obtaining maintenance costs based on a damage cost model: taking blade crack situation D as an example. j and maintenance measures a i Using the variable as the independent variable, construct the maintenance cost function C(a) under various combinations of crack conditions and maintenance measures. i D j ), j=1, 2, 3, i=1, 2, 3;

[0010] 3) Minimum cost decision analysis using decision tree based on Bayesian update theory: Input the blade damage probability model and damage cost model into the decision tree, and calculate the maintenance cost for the maintenance decision process before and after adding the structural health monitoring system; take the difference between the maintenance costs before and after to obtain the economic benefits brought by adding the health monitoring system.

[0011] Furthermore, the blade crack damage probability modeling process in step 1) specifically includes:

[0012] 11) There are two critical thresholds for helicopter rotor blade crack length: a lower initial crack threshold and an upper critical crack threshold. The lower initial crack threshold refers to the minimum detectable crack length l0. When the crack length l is less than the minimum crack length l0, the rotor blade is considered to be in a healthy state D1. The upper critical crack threshold refers to the maximum crack length l that allows the rotor blade to maintain its integrity. c The crack length l is greater than the maximum crack length l c At that time, the blade was considered to be in an irreparable state D3; therefore, the probabilities P(D) of the three crack scenarios are... j ) is represented as:

[0013] P(D1) = P{l(t) < l0}

[0014] P(D2)=P{l0≤l(t)<l c}

[0015] P(D3)=P{l(t)≥l c}

[0016] In the formula, l(t) is the crack length at time t, and P{·} is the probability value;

[0017] 12) The probability P(D1) that the blade is in a healthy, crack-free state is the blade's reliability R(t0) at that moment; blade failure follows a Weibull failure distribution, and its reliability formula is:

[0018]

[0019] In the formula, t is the time variable, λ > 0 is the proportional parameter, k > 0 is the blade shape parameter, and e is a constant; We get:

[0020] P(D1) = R(t0);

[0021] 13) For cracks with a length greater than the maximum crack length l c Using structural reliability modeling methods, the probability P(D3) of the blade crack being an irreparable state D3 is calculated; the critical function g F (t) represents the maximum crack length l c The difference between the crack length l(t) at time t and the crack length at time t:

[0022] g F (t)=l c -l(t)

[0023] According to the Paris crack propagation formula, the crack propagation rate is:

[0024]

[0025] Where l is the crack length, n is the number of cycles, ΔS is the stress range of a single cycle, and c and m are model parameters; n = vt, v is the annual cycle rate, and t is in years; with an initial crack length of l0, solving the differential equation of crack propagation rate yields l(t), a function of crack length l versus time t. Therefore, the crack length is greater than the maximum crack length l. c The probability P(D3) is given by the critical function g at time t0. F The probability that (t0) is not greater than 0:

[0026] P(D3)={g F (t0)≤0};

[0027] The three crack scenarios are independent, and their probabilities sum to 1. Therefore, the probability that the blade is in a repairable state (D2) is:

[0028] P(D2) = 1 - P(D1) - P(D3).

[0029] Furthermore, the process of obtaining repair costs based on the damage cost model in step 2) specifically includes:

[0030] 21) Maintenance costs consist of blade repair costs and risk costs, with blade repair costs C m Only with maintenance measures a i Related to, and risk cost C risk Maintenance measure a i And blade crack condition D j Jointly decided; for different maintenance measures a i The cost of blade repair is further divided into: cost of a single blade C s Condition-based maintenance cost of propeller blades after t0 flight hours (C) r :

[0031]

[0032] 22) Risk Cost C risk This refers to the cost incurred because the repair measures taken cannot restore the damaged blades to a reliable and safe operating level, thus exposing the rotor to failure risk. This is the additional cost incurred when i < j, excluding the blade repair cost. The worst-case scenario is replacing the entire rotor system; other risk costs equal the repair cost of further damage.

[0033]

[0034] 23) Maintenance cost function C(a i D j ) = C m +C risk The details are as follows:

[0035]

[0036] Among them, C rotor The cost of the rotor system is equal to the sum of the production costs of the four blades and the hub.

[0037] Furthermore, step 3) specifically includes using a decision tree based on Bayesian update theory to perform minimum cost decision analysis, which specifically includes:

[0038] 31) Before adding a structural health monitoring system, for maintenance measures a i Cost C(a) i The probability P(D) depends on the initial crack condition. j ):

[0039]

[0040] Choose the action with the lowest cost. opt The minimum maintenance cost C is obtained by decision tree analysis with the objective of minimizing maintenance costs. min :

[0041]

[0042]

[0043] 32) When a structural health monitoring system is available, the vibration information y obtained by monitoring the blades is used. k This reduces the uncertainty of the blade condition, specifically the probability P(D) of the original three crack scenarios. j Perform a Bayesian update and obtain the optimization probability P(D) based on the current actual condition of the blades. j |y k The frequency of the propeller blades gradually decreases as they transition from a healthy state (D1) to a repairable state (D2) and then to an unrepairable state (D3); when y is detected... k In the case of blade cracking, condition D j The optimization probability is:

[0044]

[0045] 33) The probability P(D) of the original blade crack situation. jReplace ) with the optimized probability P(D) of blade cracking. j |y k Using the same decision tree method, the minimum maintenance cost after adding the structural health monitoring system was obtained. for:

[0046]

[0047] The value of the structural health monitoring system is thus quantified as the difference between the minimum maintenance cost before and after implementing the system, i.e., the maximum value (VoI) that needs to be paid.

[0048]

[0049] In the formula, E y (·) is the expected value of y on the distribution of y, P(y k ) is the monitoring frequency y k The probability value.

[0050] This invention also provides an information value quantification terminal, comprising:

[0051] One or more processors;

[0052] Memory, used to store one or more programs;

[0053] When the one or more programs are executed by the one or more processors, the one or more processors implement the information value quantification method.

[0054] The beneficial effects of this invention are:

[0055] This invention quantifies the value of structural health monitoring information by using a damage probability model, a decision tree method based on Bayesian theory, and a system cost and benefit analysis model to model structural performance and utility. This solves the problem of the difficulty in quantifying the economics of structural health monitoring, and is of great significance for operators to judge the impact on helicopter operations before investment and to promote the application of structural health monitoring. Attached Figure Description

[0056] Figure 1 This is a flowchart illustrating the method of the present invention.

[0057] Figure 2a This is a schematic diagram of decision tree decision analysis for an unstructured health monitoring system.

[0058] Figure 2b A schematic diagram of decision tree analysis for incorporating structural health monitoring systems.

[0059] Figure 3 Let y be the assumed frequency offset of the monitored frequency compared to the frequency of a healthy leaf. kThe probability P(D) of the relationship between crack state and frequency when the crack rate is 1.25%. j |y k (Diagram)

[0060] Figure 4 This is a schematic diagram illustrating the probability of crack conditions after optimization based on Bayesian theory and monitoring information.

[0061] Figure 5 This is a schematic diagram showing the monitoring frequency versus cost curves under different crack conditions. Detailed Implementation

[0062] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to embodiments and accompanying drawings. The content mentioned in the embodiments is not intended to limit the present invention.

[0063] Reference Figure 1 As shown, the present invention provides a method for quantifying the value of structural health monitoring information for helicopter rotor blades, comprising the following steps:

[0064] 1) Establish a blade damage probability model based on Weibull failure distribution and Paris crack propagation formula: classify helicopter blade crack conditions into healthy state D1, repairable state D2, and unrepairable state D3; for the three crack conditions, there are three corresponding maintenance measures: no maintenance a1, repair crack a2, and blade replacement a3; according to the blade Weibull failure distribution and Paris crack propagation formula, perform probability modeling for the three crack conditions respectively.

[0065] The blade crack damage probability modeling process specifically includes:

[0066] 11) There are two critical thresholds for helicopter rotor blade crack length: a lower initial crack threshold and an upper critical crack threshold. The lower initial crack threshold refers to the minimum detectable crack length l0. When the crack length l is less than the minimum crack length l0, the rotor blade is considered to be in a healthy state D1. The upper critical crack threshold refers to the maximum crack length l that allows the rotor blade to maintain its integrity. c The crack length l is greater than the maximum crack length l c At that time, the blade was considered to be in an irreparable state D3; therefore, the probabilities P(D) of the three crack scenarios are... j ) is represented as:

[0067] P(D1) = P{l(t) < l0}

[0068] P(D2)=P{l0≤l(t)<l c}

[0069] P(D3)=P{l(t)≥l c}

[0070] In the formula, l(t) is the crack length at time t, and P{·} is the probability value;

[0071] 12) The probability P(D1) that the blade is in a healthy, crack-free state is the reliability R(t0) of the blade at this point; the blade failure follows a Weibull failure distribution, and its reliability formula is:

[0072]

[0073] In the formula, t is the time variable, λ > 0 is the proportional parameter, k > 0 is the blade shape parameter, and e is a constant; We get:

[0074] P(D1) = R(t0);

[0075] 13) For cracks with a length greater than the maximum crack length l c Using structural reliability modeling methods, the probability P(D3) of the blade crack being an irreparable state D3 is calculated; the critical function g F (t) represents the maximum crack length l c The difference between the crack length l(t) at time t and the crack length at time t:

[0076] g F (t)=l c -l(t)

[0077] According to the Paris crack propagation formula, the crack propagation rate is:

[0078]

[0079] Where l is the crack length, n is the number of cycles, ΔS is the stress range of a single cycle, and c and m are model parameters; n = vt, v is the annual cycle rate, and t is in years; with an initial crack length of l0, solving the differential equation of crack propagation rate yields l(t), a function of crack length l versus time t. Therefore, the crack length is greater than the maximum crack length l. c The probability P(D3) is given by the critical function g at time t0. F The probability that (t0) is not greater than 0:

[0080] P(D3)={g F (t0)≤0};

[0081] The three crack scenarios are independent, and their probabilities sum to 1. Therefore, the probability that the blade is in a repairable state (D2) is:

[0082] P(D2) = 1 - P(D1) - P(D3).

[0083] 2) Obtaining maintenance costs based on a damage cost model: taking blade crack situation D as an example. jand maintenance measures a i Using the variable as the independent variable, construct the maintenance cost function C(a) under various combinations of crack conditions and maintenance measures. i D j ), j=1, 2, 3, i=1, 2, 3;

[0084] The process of obtaining repair costs based on the damage cost model specifically includes:

[0085] 21) Maintenance costs consist of blade repair costs and risk costs, with blade repair costs C m Only with maintenance measures a i Related to, and risk cost C risk Maintenance measure a i And blade crack condition D j Jointly decided; for different maintenance measures a i The cost of blade repair is further divided into: cost of a single blade C s Condition-based maintenance cost of propeller blades after t0 flight hours (C) r :

[0086]

[0087] 22) Risk Cost C risk This refers to the cost incurred because the repair measures taken cannot restore the damaged blades to a reliable and safe operating level, thus exposing the rotor to failure risk. This is the additional cost incurred when i < j, excluding the blade repair cost. The worst-case scenario is replacing the entire rotor system; other risk costs equal the repair cost of further damage.

[0088]

[0089] 23) Maintenance cost function C(a i D j ) = C m +C risk The details are as follows:

[0090]

[0091] Among them, C rotor The cost of the rotor system is equal to the sum of the production costs of the four blades and the hub.

[0092] 3) Minimum cost decision analysis using decision tree based on Bayesian update theory: Input the blade damage probability model and damage cost model into the decision tree, and calculate the maintenance cost for the maintenance decision process before and after adding the structural health monitoring system; take the difference between the maintenance costs before and after to obtain the economic benefits brought by adding the health monitoring system.

[0093] Specifically, the minimum cost decision analysis based on decision trees using Bayesian update theory includes:

[0094] 31) Before adding a structural health monitoring system, for maintenance measures a i Cost C(a) i The probability P(D) depends on the initial crack condition. j ):

[0095]

[0096] Choose the action with the lowest cost. opt The minimum maintenance cost C is obtained by decision tree analysis with the objective of minimizing maintenance costs. min :

[0097]

[0098]

[0099] 32) When a structural health monitoring system is available, the vibration information y obtained by monitoring the blades is used. k This reduces the uncertainty of the blade condition, specifically the probability P(D) of the original three crack scenarios. j Perform a Bayesian update and obtain the optimization probability P(D) based on the current actual condition of the blades. j |y k The frequency of the propeller blades gradually decreases as they transition from a healthy state (D1) to a repairable state (D2) and then to an unrepairable state (D3); when y is detected... k In the case of blade cracking, condition D j The optimization probability is:

[0100]

[0101] 33) The probability P(D) of the original blade crack situation. j Replace ) with the optimized probability P(D) of blade cracking. j |y k Using the same decision tree method, the minimum maintenance cost after adding the structural health monitoring system was obtained. for:

[0102]

[0103] The value of the structural health monitoring system is thus quantified as the difference between the minimum maintenance cost before and after implementing the system, i.e., the maximum value (VoI) that needs to be paid.

[0104]

[0105] In the formula, Ey (·) is the expected value of y on the distribution of y, P(y k ) is the monitoring frequency y k The probability value.

[0106] The process of using decision tree method to perform cost analysis on the maintenance decision-making process before and after the addition of the structural health monitoring system is as follows: Figure 2a , Figure 2b As shown. Without a structural health monitoring system, P(D) j The damage probability model is obtained from (1) based on the Weibull failure distribution and the Paris crack propagation model. Taking maintenance measure a2 as an example, the blade crack condition D j The corresponding cost is C(a2,D) j ), j=1,2,3, the total cost of maintenance measure a2 is ∑C(a2,D j )P(D j The minimum cost C for the decision optimization objective. min After selecting to join this monitoring system, the monitoring frequency {y} reduces the uncertainty of crack conditions. This is achieved through Bayesian theory and the monitoring value y. k To obtain a more realistic probability P(D) of blade cracking. j |y k The natural frequency of a blade can reflect its structural state. Vibration-based structural health monitoring analyzes long-term monitoring of blade vibration frequency data and extracts features to obtain the relationship between crack condition and frequency, P(D). j |y k ),like Figure 3 As shown. When y was detected... k In the case of crack condition D j The optimization probability P(D) j |y k ),like Figure 4 As shown. The minimum cost can be obtained from the optimized blade crack probability.

[0107] The significance of the information-value-based optimized maintenance decision-making framework proposed in this invention lies in predicting the cost savings that a structural health monitoring system can save before its implementation. k These are values ​​measured after the system is put into monitoring. To achieve the goal of this invention, the frequency value is changed from discrete y k Generalization to continuous {y} is necessary, such as Figure 5 As shown. When maintenance measures a are taken i When the expected value of y is calculated on the distribution curve, the minimum maintenance cost is obtained. Thus, the value of the structural health monitoring system is quantified as the difference between the minimum maintenance costs before and after its implementation, i.e., the highest price the operator is willing to pay for the system.

[0108] Furthermore, the present invention also provides an information value quantification terminal, comprising:

[0109] One or more processors;

[0110] Memory, used to store one or more programs;

[0111] When the one or more programs are executed by the one or more processors, the one or more processors implement the information value quantification method.

[0112] This invention has many specific applications. The above description is only a preferred embodiment of this invention. It should be noted that for those skilled in the art, several improvements can be made without departing from the principle of this invention, and these improvements should also be considered within the scope of protection of this invention.

Claims

1. A method for quantifying the value of structural health monitoring information for helicopter rotor blades, characterized in that, The steps are as follows: 1) Establish a blade damage probability model based on Weibull failure distribution and Paris crack propagation formula: classify helicopter blade crack conditions into healthy state D1, repairable state D2, and unrepairable state D3; for the three crack conditions, there are three corresponding maintenance measures: no maintenance a1, repair crack a2, and blade replacement a3; according to the blade Weibull failure distribution and Paris crack propagation formula, perform probability modeling for the three crack conditions respectively. 2) Obtaining maintenance costs based on a damage cost model: taking blade crack situation D as an example. j and maintenance measures a i Using the variable as the independent variable, construct the maintenance cost function C(a) under various combinations of crack conditions and maintenance measures. i D j ), j=1,2,3, i=1,2,3; 3) Minimum cost decision analysis using decision trees based on Bayesian update theory: Input the blade damage probability model and damage cost model into the decision tree, and calculate the maintenance cost for the maintenance decision process before and after adding the structural health monitoring system; take the difference between the maintenance costs before and after. The blade crack damage probability modeling process in step 1) specifically includes: 11) There are two critical thresholds for helicopter rotor blade crack length: a lower initial crack threshold and an upper critical crack threshold; the lower initial crack threshold refers to the minimum detectable crack length. Crack length Less than the minimum crack length The blade is considered to be in a healthy state (D1); the upper critical crack threshold refers to the maximum crack length that allows the blade to maintain its integrity. Crack length Greater than the maximum crack length At that time, the blade was considered to be in an irreparable state D3; therefore, the probabilities P(D) of the three crack scenarios are... j ) is represented as: ； ； ； In the formula, Let be the crack length at time t. This is a probability value; 12) The probability P(D1) that the blade is in a healthy, crack-free state is the reliability R(t0) of the blade at this point; blade failure follows a Weibull failure distribution, and its reliability formula is: ； In the formula, t is the time variable, λ > 0 is the proportional parameter, k > 0 is the blade shape parameter, and e is a constant; We get: ； 13) For cracks with a length greater than the maximum crack length Using structural reliability modeling methods, the probability P(D3) of the blade crack being an irreparable state D3 is calculated; critical function. Represented as maximum crack length Crack length at time t The difference: ； According to the Paris crack propagation formula, the crack propagation rate is: ； in, Where n is the crack length and n is the number of cycles. This refers to the single-cycle stress range, where c and m are model parameters; n = vt, where v is the annual cycle rate and t is in years; the initial crack length is... In this case, the crack length can be obtained by solving the differential equation of crack propagation rate. function of time t Therefore, the crack length is greater than the maximum crack length. The probability P(D3) is the critical function at time t0. The probability is not greater than 0: ； The three crack scenarios are independent, and their probabilities sum to 1. Therefore, the probability that the blade is in a repairable state (D2) is: 。 2. The method for quantifying the value of structural health monitoring information for helicopter rotor blades according to claim 1, characterized in that, The process of obtaining repair costs based on the damage cost model in step 2) specifically includes: 21) Maintenance costs consist of blade repair costs and risk costs, with blade repair costs C m Only with maintenance measures a i Related to, and risk cost C risk Maintenance measure a i And blade crack condition D j Jointly decided; for different maintenance measures a i The cost of blade repair is further divided into: cost of a single blade C s Condition-based maintenance cost of propeller blades after t0 flight hours (C) r : ； 22) Risk Cost C risk This refers to the cost incurred because the repair measures taken could not restore the damaged blades to a reliable and safe operating level, resulting in the rotor being exposed to failure risk. This is the additional cost incurred when i < j, excluding the blade repair cost. The result is the replacement of the entire rotor system; other risk costs equal the repair cost for further damage. ； 23) Maintenance cost function C(a) i D j )= C m + C risk The details are as follows: ； Among them, C rotor The cost of the rotor system is equal to the sum of the production costs of the four blades and the hub.

3. The method for quantifying the value of structural health monitoring information for helicopter rotor blades according to claim 2, characterized in that, Step 3) specifically includes using decision trees based on Bayesian update theory to perform minimum cost decision analysis, which includes: 31) Before adding a structural health monitoring system, for maintenance measures a i Cost C(a) i The probability P(D) depends on the initial crack condition. j ): ； Choose the action with the lowest cost. opt The minimum maintenance cost is obtained through decision tree analysis with the objective of minimizing maintenance costs. : ； ； 32) When a structural health monitoring system is available, the vibration information y obtained by monitoring the blades... k This reduces the uncertainty of the blade condition, specifically the probability P(D) of the original three crack scenarios. j Perform a Bayesian update to obtain the optimization probability based on the current condition of the blades. The frequency of the propeller blades gradually decreases as they transition from a healthy state (D1) to a repairable state (D2) and then to an unrepairable state (D3); when y is detected... k In the case of blade cracking, condition D j The optimization probability is: ； 33) The probability P(D) of the original blade crack situation. j Replace ) with the optimized blade crack probability P(D) j | y k Using the same decision tree method, the minimum maintenance cost after adding the structural health monitoring system was obtained. for: ； Thus, the value of the structural health monitoring system is quantified as the difference between the minimum maintenance cost before and after implementing the system, i.e., the maximum cost that needs to be paid. : ； In the formula, It involves finding the expected value on the distribution of y. The monitoring frequency is The probability value.

4. An information value quantification terminal, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the information value quantification method as described in any one of claims 1-3.

Images