Method and device for determining fault probability of fan system, storage medium and electronic device

Abstract

The invention discloses a method and device for determining the fault probability of a fan system, a storage medium and an electronic device.The method comprises the steps that first fault data are obtained from historical operation records of the fan system, and second fault data from a target object are received, the first fault data represents data of a fault event of the fan system, and the second fault data represents data of risk assessment of the fault event of the fan system by the target object; determining a fault reason of the fault event from the first fault data and the second fault data, and determining a risk factor of the fan system according to the fault reason; and determining the fault probability according to the prior probability of the risk factor and the objective weight of the risk factor. By adopting the technical scheme, the problem that the fault probability of the fan system cannot be accurately determined is solved.

Description

Technical Field

[0001] The present application relates to the field of determining the failure probability of a wind turbine system, and specifically, to a method, device, storage medium, and electronic device for determining the failure probability of a wind turbine system. Background Art

[0002] In wind farms, failures in wind turbine systems can result in significant economic losses. Therefore, effective assessment of wind turbine failure risks is necessary to improve wind farm reliability. Currently, wind power companies typically conduct surveys and statistics on wind turbine failures, using these data to develop failure models and predict the probability of wind turbine failure.

[0003] However, this method generally uses historical data to establish a wind turbine failure model, without considering the impact of the wind farm's operating conditions on the failure probability. The established failure model is not accurate enough, resulting in the assessed failure probability being unable to reflect the current status of the wind turbine, and therefore unable to accurately determine the failure probability of the wind turbine system.

[0004] Currently, no effective solution has been proposed to the problem in the related art that the failure probability of the fan system cannot be accurately determined. Therefore, it is necessary to improve the related art to overcome the above-mentioned drawbacks. Summary of the Invention

[0005] Embodiments of the present application provide a method, device, storage medium, and electronic device for determining the failure probability of a wind turbine system, so as to at least solve the problem of being unable to accurately determine the failure probability of a wind turbine system.

[0006] According to one aspect of an embodiment of the present application, a method for determining the failure probability of a wind turbine system is provided, comprising: obtaining first fault data from a historical operation record of the wind turbine system, and receiving second fault data from a target object, wherein the first fault data represents data of a failure event of the wind turbine system, and the second fault data represents data of a risk assessment of the failure event of the wind turbine system by the target object; determining a fault cause of the failure event from the first fault data and the second fault data, and determining a risk factor of the wind turbine system based on the fault cause; and determining the failure probability based on the prior probability of the risk factor and the objective weight of the risk factor.

[0007] In an exemplary embodiment, determining the risk factor of the wind turbine system according to the fault cause includes at least one of the following: when it is determined that the type of the fault event is a mechanical fault, locating the location of the mechanical fault, and determining the mechanical fault risk factor according to the location and the mechanical fault cause of the mechanical fault; when it is determined that the type of the fault event is an electrical fault, determining the electrical fault cause corresponding to the electrical fault, and obtaining the electrical fault risk factor corresponding to the electrical fault cause; determining the mechanical failure risk factor and the electrical failure risk factor as the risk factor.

[0008] In an exemplary embodiment, before determining the failure probability based on the prior probability of the risk factor and the objective weight of the risk factor, the method further includes: calculating the weight entropy of the risk factor using the subjective judgment level vector of the risk factor, and calculating the objective weight using the weight entropy; the weight entropy is expressed as H k , then the objective weight a is calculated using the weight entropy using the following formula k :

[0009] m is a positive integer.

[0010] In an exemplary embodiment, before determining the failure probability based on the prior probability of the risk factor and the objective weight of the risk factor, the prior probability is determined in the following manner: obtaining the current operating data of the wind turbine system; determining the prior probability based on the weighted sum of the failure probability corresponding to the first fault data, the failure probability corresponding to the second fault data, and the failure probability corresponding to the current operating data; or determining the prior probability based on the result of fuzzy comprehensive evaluation of the failure probability corresponding to the first fault data, the failure probability corresponding to the second fault data, and the failure probability corresponding to the current operating data.

[0011] In an exemplary embodiment, the failure probability is determined based on the prior probability of the risk factor and the objective weight of the risk factor, including: determining the membership parameter corresponding to the judgment level of the prior probability from the membership parameter table of the risk factor, the membership parameter table storing the judgment result of the risk factor, the judgment level, the membership parameter, and the value range of the prior probability; determining the target membership parameter based on the weighted sum value between the objective weight and the membership parameter, and determining the target membership parameter as the failure probability.

[0012] In an exemplary embodiment, the method further includes: obtaining a Bayesian network constructed based on the risk factor and the fault event, the root node of the Bayesian network represents the risk factor, and the risk factor has a prior probability, the leaf node represents the fault event, and there is an intermediate node between the root node and the leaf node; calculating the prior probability of the leaf node based on the prior probability of the root node and the conditional probability of all nodes in the Bayesian network, and determining the fault probability based on the prior probability.

[0013] According to another aspect of an embodiment of the present application, a device for determining the failure probability of a wind turbine system is also provided, including: an acquisition module for acquiring first fault data from the historical operation record of the wind turbine system, and receiving second fault data from a target object, wherein the first fault data represents data of a failure event of the wind turbine system, and the second fault data represents data of a risk assessment of the failure event of the wind turbine system by the target object; a first determination module for determining the cause of the failure event from the first fault data and the second fault data, and determining the risk factor of the wind turbine system based on the cause of the failure; and a second determination module for determining the failure probability based on the prior probability of the risk factor and the objective weight of the risk factor.

[0014] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored. The computer program is configured to execute the above-mentioned method for determining the failure probability of the wind turbine system when running.

[0015] According to another aspect of an embodiment of the present application, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method for determining the failure probability of the wind turbine system through the computer program.

[0016] According to another aspect of the present application, a computer program product is provided, including a computer program, which implements the steps in any of the above method embodiments when executed by a processor.

[0017] Through this application, first fault data is obtained from the historical operation records of the wind turbine system, and second fault data is received from a target object. The first fault data represents data on the wind turbine system's fault event, and the second fault data represents data on the target object's risk assessment of the wind turbine system's fault event. The cause of the fault event is determined from the first and second fault data, and a risk factor for the wind turbine system is determined based on the fault cause. The fault probability is determined based on the prior probability of the risk factor and the objective weight of the risk factor. By integrating the historical operation records, the target object's risk assessment data, and the current operation data, the method not only improves the accuracy of fault prediction but also provides a scientific basis for the maintenance and management of the wind turbine system, thereby improving the operational efficiency and safety of the wind farm. By introducing the objective weight and prior probability of the risk factor, the method can more comprehensively consider various possible fault causes, making the determination of the fault probability more reasonable and reliable. Therefore, the method solves the problem of being unable to accurately determine the fault probability of the wind turbine system, improves the accuracy of the fault probability, and achieves the effect of accurately determining the fault probability of the wind turbine system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0020] Figure 1 This is a hardware structure block diagram of a mobile terminal for a method for determining a failure probability of a wind turbine system according to an embodiment of the present application;

[0021] Figure 2 is a flow chart of a method for determining a failure probability of a wind turbine system according to an embodiment of the present application;

[0022] Figure 3 is a schematic diagram of a Bayesian network according to an embodiment of the present application;

[0023] Figure 4 This is a structural block diagram of a device for determining the failure probability of a wind turbine system according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings 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 in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal of a wind turbine system, a server device of a wind turbine system, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for determining the failure probability of a wind turbine system according to an embodiment of the present application. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a microprocessor or a processing device such as a field programmable gate array (FPGA)) and a memory 104 for storing data. The mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0027] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for determining the failure probability of the wind turbine system in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, implementing the above-mentioned method. The memory 104 may include a high-speed random access memory and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely located relative to the processor 102, and these remote memories may be connected to the mobile terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0028] The transmission device 106 is used to receive or send data via a network. A specific example of the aforementioned network may include a wireless network provided by the mobile terminal's communications provider. In one embodiment, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0029] In this embodiment, a method for determining the failure probability of a wind turbine system is provided. Figure 2 is a flow chart of a method for determining the failure probability of a wind turbine system according to an embodiment of the present application, such as Figure 2 As shown, the process includes the following steps:

[0030] Step S202: Acquire first fault data from the historical operation record of the wind turbine system, and receive second fault data from a target object, wherein the first fault data represents data of a fault event of the wind turbine system, and the second fault data represents data of a risk assessment performed by the target object on the fault event of the wind turbine system;

[0031] It should be noted that the primary fault data in historical operation records includes, but is not limited to, the type, frequency, and duration of past faults in the wind turbine system, as well as the specific conditions that led to the faults, such as excessive temperature, abnormal wind speed, and component wear. The secondary fault data for the target object can be derived from expert assessments, industry standards, or risk analysis under specific environmental conditions. For example, experts can assess the risk level of a specific fault based on the wind conditions and climate conditions in the area where the wind turbine is located, as well as the wind turbine's maintenance records. Risk factors can be derived from a comprehensive analysis of these fault causes, including the degree of mechanical wear, electrical system aging, and the impact of environmental factors.

[0032] Step S204: determining a fault cause of the fault event from the first fault data and the second fault data, and determining a risk factor of the wind turbine system according to the fault cause;

[0033] Step S206: determining the failure probability according to the prior probability of the risk factor and the objective weight of the risk factor.

[0034] It should be noted that after the risk factors are graded, the likelihood of the occurrence of the corresponding fault event can be used as the fault probability, which can reduce the amount of calculation of the joint probability distribution and thus simplify the calculation process.

[0035] Through the above steps, first fault data is obtained from the historical operation records of the wind turbine system, and second fault data is received from a target object. The first fault data represents data on the wind turbine system's fault event, and the second fault data represents data on the target object's risk assessment of the wind turbine system's fault event. The cause of the fault event is determined from the first and second fault data, and a risk factor for the wind turbine system is determined based on the fault cause. The fault probability is determined based on the prior probability of the risk factor and the objective weight of the risk factor. By integrating historical operation records, the target object's risk assessment data, and current operation data, the method not only improves the accuracy of fault prediction but also provides a scientific basis for wind turbine system maintenance and management, thereby enhancing the operational efficiency and safety of wind farms. By introducing the objective weights and prior probabilities of risk factors, the method can more comprehensively consider various possible fault causes, making the determination of fault probability more reasonable and reliable. Therefore, the method solves the problem of being unable to accurately determine the fault probability of a wind turbine system, improves the accuracy of the fault probability, and achieves the effect of accurately determining the fault probability of a wind turbine system.

[0036] By integrating historical operating records and risk assessment data of the target object, the above embodiment not only takes into account the past performance of the wind turbine system, but also incorporates the assessment of the current environment and operating conditions, thereby more accurately predicting the future failure probability of the wind turbine system. This method solves the limitations of traditional fault prediction that relies solely on historical data or a single evaluation standard, improves the accuracy and reliability of the prediction, and is of great value for the formulation of maintenance strategies for wind farms. For example, by analyzing historical fault data and assessment data of the target object, this application identifies that abnormal wind speed and component wear are the main risk factors leading to wind turbine failures, and then adjusts the maintenance plan, giving priority to inspecting and maintaining components that are more affected by wind speed, effectively preventing failures, reducing downtime, and improving the economic benefits and operational safety of wind farms.

[0037] In an exemplary embodiment, the method for determining the risk factor of the fan system based on the fault cause includes at least one of the following: when it is determined that the type of the fault event is a mechanical fault, locating the occurrence location of the mechanical fault, and determining the mechanical fault risk factor based on the occurrence location and the mechanical fault cause of the mechanical fault; when it is determined that the type of the fault event is an electrical fault, determining the electrical fault cause corresponding to the electrical fault, and obtaining the electrical fault risk factor corresponding to the electrical fault cause; and determining the mechanical failure risk factor and the electrical failure risk factor as the risk factor.

[0038] like Figure 3 As shown, mechanical failures specifically include blade breakage and deformation, bearing failure, and belt loosening and wear. The above-mentioned mechanical failure risk factors may include quality issues, design issues, manufacturing and installation issues, natural factors, poor lubrication, and service life. Among them, the failure at the blade corresponds to blade breakage and deformation, and the corresponding mechanical failure risk factors include quality issues, design issues, manufacturing and installation issues, natural factors, and service life. The failure at the bearing corresponds to bearing failure, and the corresponding mechanical failure risk factors include manufacturing and installation issues, natural factors, poor lubrication, and service life. The failure at the belt corresponds to belt loosening and wear, and the corresponding mechanical failure risk factors include natural factors, poor lubrication, and service life. Based on this, when historical data shows that blade cracks frequently appear in a specific location, the mechanical failure risk factor of that location can be set to a higher weight.

[0039] Electrical failures specifically include motor failures and electrical connection failures. Risk factors for electrical failures can include coil burnout, insulation aging, circuit failures, and poor contact. Corresponding risk factors for motor failures include coil burnout, insulation aging, circuit failures, and poor contact. Corresponding risk factors for electrical connection failures include circuit failures and poor contact.

[0040] Based on the above embodiments, different types of fault risks in wind turbine systems can be more accurately identified and quantified. For example, after the system identifies blade cracks and cable aging as major risk factors, it can specifically strengthen the monitoring and maintenance of these components. At the same time, by analyzing electrical fault risk factors, software upgrades or replacement of vulnerable components can be performed in advance to avoid sudden downtime caused by electrical faults. This approach solves the problems of inaccurate fault cause identification and irrational resource allocation in traditional maintenance strategies. Through refined risk factor management, it improves the operational stability and maintenance efficiency of wind turbine systems.

[0041] In an exemplary embodiment, before determining the failure probability based on the prior probability of the risk factor and the objective weight of the risk factor, the weight entropy of the risk factor can be further calculated using the subjective judgment level vector of the risk factor, and the objective weight can be calculated using the weight entropy; the weight entropy is expressed as H k , then the objective weight a is calculated using the weight entropy using the following formula k :

[0042] m is a positive integer.

[0043] The subjective judgment level vector reflects the experts' subjective assessments of different risk factors, while the weighted entropy quantifies the uncertainty of these assessments. For example, if the experts' assessments of gearbox wear are highly consistent, the weighted entropy will be low, meaning that the objective weight of this risk factor will be more dependent on historical data and current operating conditions. Conversely, if there is significant disagreement, the weighted entropy will be high, and the objective weight will take into account the target object's secondary fault data.

[0044] Optionally, in this embodiment, the calculation process of the objective weight may be understood by referring to the following process:

[0045] Based on the dependency between risk factors and failure probabilities, a Bayesian network is constructed. Directed edges between nodes in the Bayesian network represent causal relationships. A conditional probability table is then constructed to quantitatively describe the strength of the relationships between nodes. Risk factors in the Bayesian network are manually scored.

[0046] Next, the entropy weight method is used to improve the objectivity of the evaluation process. In this process, the natural number scale (e 0 / 5 ~e 8 / 5 ) replaces the conventional scale (1 to 9) to improve the accuracy of judgment.

[0047] Assume that the result of the sth expert is the optimal result, set as W s =(W 1s ,W 2s ,...,W ns ), the remaining experts make subjective comparisons with the optimal evaluation results to determine the pros and cons of their own evaluation results. The subjective evaluation level vector of the kth expert is expressed as E k .

[0048] E k =(e 1k ,e 2k ,...,e nk )(k=1,2,...,m).

[0049]

[0050] Then establish the weight entropy model of its own evaluation results:

[0051]

[0052] Among them, H k is the uncertainty of expert k’s subjective judgment result, H k The larger the value, the lower the credibility of the evaluation result of expert k. ik is the entropy value of the subjective judgment result of expert k on the i-th indicator.

[0053] Determine the objective weight of the kth expert based on the above weight entropy model:

[0054]

[0055] This embodiment introduces the concept of weighted entropy to scientifically calculate the objective weights of risk factors, ensuring the objectivity and accuracy of fault probability assessments. This approach addresses the problem of subjective weight assignment and a lack of scientific basis in fault probability assessments. By quantifying the uncertainty of expert assessments, it makes the objective weights of risk factors more reasonable, thereby improving the reliability of fault probability assessments. For example, a low calculated weighted entropy for gearbox wear indicates consensus among experts. Therefore, the objective weight for gearbox wear will be based more on historical fault data and current operating status. However, a high weighted entropy for blade cracks indicates divergent assessments, and its objective weight will take into account the target object's secondary fault data, ensuring the comprehensiveness and objectivity of the assessment results.

[0056] In an exemplary embodiment, before determining the failure probability based on the prior probability of the risk factor and the objective weight of the risk factor, the prior probability can also be determined in the following manner: obtaining the current operating data of the wind turbine system; determining the prior probability based on the weighted sum of the failure probability corresponding to the first fault data, the failure probability corresponding to the second fault data, and the failure probability corresponding to the current operating data; or determining the prior probability based on the result of fuzzy comprehensive evaluation of the failure probability corresponding to the first fault data, the failure probability corresponding to the second fault data, and the failure probability corresponding to the current operating data.

[0057] It should be noted that the current operating data may include real-time parameters such as temperature, wind speed, vibration, etc. These data are combined with historical fault data and the evaluation data of the target object, and the prior probability is calculated by weighted sum or fuzzy comprehensive evaluation. For example, if the current wind speed is abnormally high, and historical data shows that the wind turbine is prone to failure under high wind speed conditions, then the risk factor of abnormal wind speed will be given a higher weight, thereby increasing the evaluation value of the prior probability. This embodiment combines historical data, target object evaluation and current operating status, and adopts weighted sum or fuzzy comprehensive evaluation technology to determine the prior probability, thereby realizing a dynamic evaluation of the probability of failure of the wind turbine system. This method solves the static and lagging problems of traditional fault prediction models. By updating the prior probability in real time, it can timely reflect the current risk status of the wind turbine system and provide a scientific basis for immediate maintenance decisions of the wind farm.

[0058] In this embodiment, for example, the following topological relationship is set:

[0059] “Graph TD

[0060] A[Historical Engineering Case Library]——>D[Probabilistic Fusion Calculation]

[0061] B[Expert Experience Database]——>D

[0062] C[Real-time monitoring of data flow]——>D

[0063] D——>E[prior probability of risk factor]

[0064] D——>F[Bayesian network input]".

[0065] Use P final =aP 历史 +bP 专家 +cP 监测 Combine different data sources (historical cases, expert experience, objective data) to obtain P final , and continue to judge P final The evaluation level.

[0066] In an exemplary embodiment, a specific scheme for determining the failure probability based on the prior probability of the risk factor and the objective weight of the risk factor includes: determining the membership parameter corresponding to the evaluation level of the prior probability from the membership parameter table of the risk factor, the membership parameter table storing the evaluation result of the risk factor, the evaluation level, the membership parameter, and the value range of the prior probability; determining the target membership parameter based on the weighted sum value between the objective weight and the membership parameter, and determining the target membership parameter as the failure probability.

[0067] It should be noted that risk factor evaluation levels can include low risk, medium risk, and high risk, and the membership parameter reflects the degree of correlation between the risk factor and the evaluation level. For example, when the prior probability falls within the high risk range, the corresponding high risk evaluation level membership parameter will be assigned a higher value, indicating that the wind turbine system has a higher failure risk in its current state.

[0068] By introducing the concepts of membership parameters and evaluation levels, the above-mentioned method achieves a refined assessment of the prior probability of risk factors, thereby more accurately determining the probability of failure. This method solves the problem of the rough assessment of risk factors and the inability to accurately reflect the risk status in traditional fault probability assessment. Through the membership parameter table, it is possible to quickly locate the corresponding evaluation level based on the value range of the prior probability, and then calculate the failure probability. For example, based on the prior probability, it is determined that the wind turbine system is in a high-risk state. By calculating the membership parameters of the high-risk evaluation level, it is concluded that the failure probability is high. This provides a clear risk warning for the wind farm maintenance team, guiding them to prioritize high-risk wind turbines, effectively preventing failures, reducing economic losses, and improving the operational efficiency and safety of the wind farm.

[0069] Furthermore, the process of determining the evaluation level can be described in Table 1 below.

[0070] Determine the prior probability P of the risk factor with the help of historical cases (corresponding to the first failure data), expert subjective experience (corresponding to the second failure data) and objective data (corresponding to the current year's operation data) final , and then the risk factors are divided into risk levels, where the prior probability distribution can also be expressed as the degree of membership of the risk factor to the evaluation level.

[0071] As shown in Table 1 below, manual evaluation is divided into seven levels. When evaluating the probability of a risk factor, two levels may overlap. If manual evaluation determines that a risk factor is very unlikely to occur or is relatively unlikely to occur, consider fuzzifying adjacent levels and determining the membership parameters using a triangular membership function.

[0072] Table 1 Membership parameters of risk factors

[0073]

[0074]

[0075] The aggregated fuzzy number of each review object (i.e., risk factor) is calculated by objective weights and manual evaluation results as follows:

[0076]

[0077] Among them, Mi is the membership parameter after aggregation of all basic events, N e is the number of people, A ij The scoring result for each person corresponds to the membership parameter, w j is the weight coefficient of each expert, and N is the total number of review objects.

[0078] w j That is, the objective weight a calculated by the entropy weight method k , which is used to aggregate the fuzzy judgment results of each expert on the probability of occurrence of risk factors. Among them, the experts with high credibility (a k The larger the score is, the greater the final membership parameter M i The impact is greater, which can effectively improve the objectivity and reliability of the prior probability calculation results.

[0079] In an optional embodiment, after obtaining the above-mentioned aggregate fuzzy number M i After that, we can use the left-right fuzzy sorting method to sort M i The fuzzy number is converted into the fuzzy probability score FPS, thus providing a data basis for subsequent calculations.

[0080] In the process of fuzzy number conversion, it is necessary to determine the intersection of the maximum fuzzy set, the minimum fuzzy set, and the function curve describing the evaluation object. The maximum fuzzy set and the minimum fuzzy set are expressed as follows:

[0081]

[0082] The intersection values are:

[0083]

[0084] Or:

[0085]

[0086] On this basis, the fuzzy feasibility score FPS corresponding to the fuzzy number is expressed as:

[0087]

[0088] Furthermore, convert FPS to FFR:

[0089]

[0090] FFR is the core parameter for converting fuzzy probability scores (FPS) into engineering probabilities, transforming fuzzy values into quantifiable probabilities. The cube root design of the K value enhances the resolution of intermediate probability intervals and improves the dynamic adjustment capability during the conversion.

[0091] Next, the interval to which the review object belongs is determined based on the FPS. Specifically, the fault event is divided into four levels (Level I is acceptable risk, Level II is tolerable risk, Level III is unacceptable risk, and Level IV is intolerable risk) based on the proportional difference between the FPS corresponding membership intervals. Level I corresponds to when the fault event does not occur.

[0092] Optionally, in one embodiment, a process for calculating the FFR value is provided in combination with the following steps:

[0093] Step 1: Calculate the FPS value. For example, the expert aggregated ratings give an FPS of 0.58.

[0094] Step 2: Calculate FFR.

[0095]

[0096] Step 3: Determine the risk level. 1.9%∈(1%,33%]→Level II tolerable risk.

[0097] Step 4: Membership correction. If the membership is at the boundary between the original level 2 (1%-10%) and level 3 (10%-33%), linear interpolation is used to obtain

[0098] Step 5: The final judgment is close to the lower limit of Level II, and it is recommended to adopt Level II basic measures + Level I monitoring.

[0099] In an exemplary embodiment, other methods for determining the probability of failure are further proposed, specifically including: obtaining a Bayesian network constructed based on the risk factor and the fault event, the root node of the Bayesian network represents the risk factor, and the risk factor has a prior probability, the leaf node represents the fault event, and there is an intermediate node between the root node and the leaf node; calculating the prior probability of the leaf node based on the prior probability of the root node and the conditional probability of all nodes in the Bayesian network, and determining the fault probability based on the prior probability. By constructing a Bayesian network, the complex relationship modeling of the fault events of the wind turbine system is achieved, and the fault probability can be evaluated more comprehensively. This method solves the problem of insufficient consideration of the mutual influence between fault events in traditional fault prediction models. Through the Bayesian network, it is possible to clearly depict how different risk factors affect the final fault event through intermediate nodes, thereby calculating a more accurate fault probability. For example, abnormal wind speed, as a risk factor, affects leaf nodes (such as wind turbine shutdown) through intermediate nodes (such as blade damage and gearbox overload). By calculating the prior probability of the root node and the conditional probability of the intermediate node, the prior probability of the leaf node is obtained, and then the failure probability of the wind turbine shutdown is determined, which provides a scientific basis for fault prevention and maintenance decisions in wind farms, and effectively improves the operating efficiency and safety of wind farms.

[0100] It's important to note that a Bayesian network is a probabilistic graphical model used to represent conditional dependencies between variables. In this approach, risk factors serve as root nodes, fault events serve as leaf nodes, and intermediate nodes represent the paths along which faults propagate or the interactions between them. For example, abnormal wind speed, as a risk factor, can lead to blade damage, which in turn can cause gearbox overload and ultimately turbine shutdown. This series of events is connected in a Bayesian network through intermediate nodes.

[0101] Obviously, the embodiments described above are only part of the embodiments of the present application, rather than all the embodiments. In order to better understand the above method, the above process is described below in conjunction with the embodiments, but it is not intended to limit the technical solutions of the embodiments of the present application. Specifically:

[0102] Furthermore, in this embodiment, the prior probability can be calculated in combination with a Bayesian network.

[0103] By combining graph theory and probability statistics, Bayesian networks can systematically reflect the logical relationship between variables within a complex system. They can not only describe the uncertainty between events and factors, but also the probability of risk factors obtained through analysis can better reflect the stability and correlation of the system.

[0104] As shown in Table 2 below, given the conditional probabilities of all nodes in a Bayesian network and the prior probability of the root node, the prior probabilities of the corresponding child nodes can be calculated using Bayes' theorem. The process of applying Bayesian networks to risk analysis involves: Based on the dependencies between risk factors, a Bayesian network topology diagram is constructed to qualitatively describe the causal relationships between nodes. Then, a conditional probability table is constructed to quantitatively describe the strength of the relationships between nodes in the network.

[0105] Table 2 Node correlation table of Bayesian network

[0106]

[0107] Taking the wind turbine gearbox system as an example, the following network topology is constructed:

[0108] “graph TD

[0109] A[natural factors]——>B[blade failure]

[0110] C[Poor lubrication]——>D[Gear box failure]

[0111] D——>E[Transmission system failure]

[0112] B——>E

[0113] E——>F[Fan system failure]

[0114] G[manufacturing defect]——>D

[0115] H[Monitoring failure]——>D".

[0116] In addition, when using the above-mentioned Bayesian network to analyze the probability of occurrence of each node under various circumstances (only considering the most likely failure events), if n nodes in the network topology diagram have 4 levels and each node has m root nodes, then the conditional probability table has a total of 4nm probability values. The amount of data for calculating the prior probability is huge. Therefore, it is considered to determine the risk probability of each level by combining the indicator weight with the indicator probability, thereby constructing a Bayesian network analysis model.

[0117] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0118] This embodiment also provides a device for determining the failure probability of a wind turbine system. This device is used to implement the above-mentioned embodiments and preferred implementations, and details already described will not be repeated. As used below, the term "module" may refer to a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0119] Figure 4 1 is a structural block diagram of a device for determining a failure probability of a wind turbine system according to an embodiment of the present application, the device comprising:

[0120] an acquisition module 42 configured to acquire first fault data from a historical operation record of the wind turbine system and receive second fault data from a target object, wherein the first fault data represents data of a fault event of the wind turbine system and the second fault data represents data of a risk assessment performed by the target object on the fault event of the wind turbine system;

[0121] A first determining module 44 is configured to determine a fault cause of the fault event from the first fault data and the second fault data, and determine a risk factor of the wind turbine system according to the fault cause;

[0122] The second determination module 46 is configured to determine the failure probability according to the prior probability of the risk factor and the objective weight of the risk factor.

[0123] The above-described device obtains first fault data from the historical operation records of the wind turbine system and receives second fault data from a target object. The first fault data represents data on the wind turbine system's fault event, and the second fault data represents data on the target object's risk assessment of the wind turbine system's fault event. The fault cause of the fault event is determined from the first and second fault data, and a risk factor for the wind turbine system is determined based on the fault cause. The fault probability is determined based on the prior probability of the risk factor and the objective weight of the risk factor. By integrating historical operation records, the target object's risk assessment data, and current operation data, the method not only improves the accuracy of fault prediction but also provides a scientific basis for wind turbine system maintenance and management, thereby enhancing the operational efficiency and safety of wind farms. By introducing the objective weight and prior probability of the risk factor, the method can more comprehensively consider various possible fault causes, making the determination of the fault probability more reasonable and reliable. Therefore, the method solves the problem of being unable to accurately determine the fault probability of a wind turbine system, improves the accuracy of the fault probability, and achieves the effect of accurately determining the fault probability of a wind turbine system.

[0124] In an exemplary embodiment, the first determination module is further used to: when it is determined that the type of the fault event is a mechanical fault, locate the location of the mechanical fault, and determine the mechanical fault risk factor based on the location and the mechanical fault cause of the mechanical fault; when it is determined that the type of the fault event is an electrical fault, determine the electrical fault cause corresponding to the electrical fault, and obtain the electrical fault risk factor corresponding to the electrical fault cause; and determine the mechanical failure risk factor and the electrical failure risk factor as the risk factor.

[0125] In an exemplary embodiment, the second determination module is further configured to, before determining the failure probability based on the prior probability of the risk factor and the objective weight of the risk factor: calculate the weight entropy of the risk factor using the subjective judgment level vector of the risk factor, and calculate the objective weight using the weight entropy; the weight entropy is expressed as H k , then the objective weight a is calculated using the weight entropy using the following formula k :

[0126] m is a positive integer.

[0127] In an exemplary embodiment, the second determination module is also used to: before determining the fault probability based on the prior probability of the risk factor and the objective weight of the risk factor, determine the prior probability in the following manner: obtain the current operating data of the wind turbine system; determine the prior probability based on the weighted sum of the fault probability corresponding to the first fault data, the fault probability corresponding to the second fault data, and the fault probability corresponding to the current operating data; or determine the prior probability based on the result of fuzzy comprehensive evaluation of the fault probability corresponding to the first fault data, the fault probability corresponding to the second fault data, and the fault probability corresponding to the current operating data.

[0128] In an exemplary embodiment, the second determination module is further used to include: determining the membership parameter corresponding to the judgment level of the prior probability from the membership parameter table of the risk factor, the membership parameter table storing the judgment result of the risk factor, the judgment level, the membership parameter, and the value range of the prior probability; determining the target membership parameter based on the weighted sum value between the objective weight and the membership parameter, and determining the target membership parameter as the failure probability.

[0129] In an exemplary embodiment, the second determination module is also used to obtain a Bayesian network constructed based on the risk factor and the fault event, the root node of the Bayesian network represents the risk factor, and the risk factor has a prior probability, the leaf node represents the fault event, and there is an intermediate node between the root node and the leaf node; the prior probability of the leaf node is calculated based on the prior probability of the root node and the conditional probability of all nodes in the Bayesian network, and the fault probability is determined based on the prior probability.

[0130] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps of any of the above method embodiments when run.

[0131] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0132] S1, obtaining first fault data from a historical operation record of the wind turbine system, and receiving second fault data from a target object, wherein the first fault data represents data of a fault event of the wind turbine system, and the second fault data represents data of a risk assessment performed by the target object on the fault event of the wind turbine system;

[0133] S2, determining a fault cause of the fault event from the first fault data and the second fault data, and determining a risk factor of the wind turbine system according to the fault cause;

[0134] S3. Determine the failure probability according to the prior probability of the risk factor and the objective weight of the risk factor.

[0135] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0136] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail here.

[0137] An embodiment of the present application further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0138] Optionally, in this embodiment, the processor may be configured to execute the following steps through a computer program:

[0139] S1, obtaining first fault data from a historical operation record of the wind turbine system, and receiving second fault data from a target object, wherein the first fault data represents data of a fault event of the wind turbine system, and the second fault data represents data of a risk assessment performed by the target object on the fault event of the wind turbine system;

[0140] S2, determining a fault cause of the fault event from the first fault data and the second fault data, and determining a risk factor of the wind turbine system according to the fault cause;

[0141] S3. Determine the failure probability according to the prior probability of the risk factor and the objective weight of the risk factor.

[0142] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0143] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0144] An embodiment of the present application further provides another computer program product, comprising a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above method embodiments are implemented.

[0145] An embodiment of the present application also provides a computer program, which includes computer instructions, which are stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps of any of the above method embodiments.

[0146] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail here.

[0147] Obviously, those skilled in the art should understand that the modules or steps of the present application described above can be implemented using a general-purpose computing device, they can be concentrated on a single computing device, or distributed across a network composed of multiple computing devices, they can be implemented using program code executable by the computing device, and thus, they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be performed in a different order than herein, or they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0148] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for determining the failure probability of a wind turbine system, characterized in that: include: Acquire first fault data from a historical operation record of the wind turbine system, and receive second fault data from a target object, wherein the first fault data represents data of a fault event of the wind turbine system, and the second fault data represents data of a risk assessment performed by the target object on the fault event of the wind turbine system; determining a fault cause of the fault event from the first fault data and the second fault data, and determining a risk factor of the wind turbine system according to the fault cause; The failure probability is determined according to the prior probability of the risk factor and the objective weight of the risk factor.

2. The method according to claim 1, characterized in that Determining a risk factor of the fan system according to the fault cause includes at least one of the following: If it is determined that the type of the fault event is a mechanical fault, locating the location of the mechanical fault, and determining a mechanical fault risk factor according to the location and the cause of the mechanical fault; In a case where it is determined that the type of the fault event is an electrical fault, determining an electrical fault cause corresponding to the electrical fault, and obtaining an electrical fault risk factor corresponding to the electrical fault cause; The mechanical failure risk factor and the electrical failure risk factor are determined as the risk factors.

3. The method according to claim 1, characterized in that Before determining the failure probability according to the prior probability of the risk factor and the objective weight of the risk factor, the method further includes: Calculating the weight entropy of the risk factor using the subjective judgment level vector of the risk factor, and calculating the objective weight using the weight entropy; The weight entropy is expressed as H k , then the objective weight a is calculated using the weight entropy using the following formula k : m is a positive integer.

4. The method according to claim 1, wherein Before determining the failure probability according to the prior probability of the risk factor and the objective weight of the risk factor, the prior probability is determined in the following manner: Get the current operating data of the fan system; Determine the prior probability according to a weighted sum of the failure probability corresponding to the first failure data, the failure probability corresponding to the second failure data, and the failure probability corresponding to the current operation data; Alternatively, the prior probability is determined based on a result of fuzzy comprehensive evaluation of the fault probability corresponding to the first fault data, the fault probability corresponding to the second fault data, and the fault probability corresponding to the current operation data.

5. The method according to claim 4, characterized in that Determining the failure probability according to the prior probability of the risk factor and the objective weight of the risk factor includes: Determining a membership parameter corresponding to the evaluation level of the prior probability from a membership parameter table of the risk factor, wherein the membership parameter table stores the evaluation result of the risk factor, the evaluation level, the membership parameter, and a value range of the prior probability; A target membership parameter is determined according to a weighted sum value between the objective weight and the membership parameter, and the target membership parameter is determined as the failure probability.

6. The method according to claim 5, characterized in that The method further comprises: Obtain a Bayesian network constructed based on the risk factor and the fault event, wherein the root node of the Bayesian network represents the risk factor, and the risk factor has a prior probability, the leaf node of the Bayesian network represents the fault event, and there is an intermediate node between the root node and the leaf node; calculate the prior probability of the leaf node according to the prior probability of the root node and the conditional probability of all nodes in the Bayesian network, and determine the fault probability according to the prior probability.

7. A device for determining the failure probability of a fan system, characterized in that: include: an acquisition module, configured to acquire first fault data from a historical operation record of the wind turbine system and receive second fault data from a target object, wherein the first fault data represents data of a fault event of the wind turbine system, and the second fault data represents data of a risk assessment performed by the target object on the fault event of the wind turbine system; a first determining module, configured to determine a fault cause of the fault event from the first fault data and the second fault data, and determine a risk factor of the wind turbine system according to the fault cause; The second determination module is configured to determine the failure probability according to the prior probability of the risk factor and the objective weight of the risk factor.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 6 when executed.

9. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 6 through the computer program.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images