Single fault tracing method and device for wind power equipment

Abstract

The invention provides a wind power equipment single fault tracing method and device, and relates to the field of wind power equipment fault analysis. The method comprises the steps of obtaining a state monitoring variable corresponding to a target subsystem in response to a fault tracing operation for wind power equipment; calculating a quantity control limit corresponding to the target subsystem through a PCA model; judging whether the wind power equipment is in a fault state or not according to quantity control limitation; if it is determined that the wind power equipment is in the fault state, obtaining training sample representation, training an SOM network structure and weight through the training sample representation, and judging whether the target subsystem is a fault subsystem or not by inputting a state monitoring variable; if it is confirmed that the target subsystem is the fault subsystem, historical fault components corresponding to the target subsystem are obtained, and the state monitoring variables and the historical fault components serve as Bayesian network nodes to construct a Bayesian network model; through a Bayesian network model, respectively calculating corresponding conditional probabilities when the target subsystem is in a plurality of fault modes; if it is confirmed that the first conditional probability is larger than the second conditional probability, the fault mode corresponding to the first conditional probability serves as a fault source corresponding to the wind power equipment, and the first conditional probability and the second conditional probability are any two different probabilities in the multiple different conditional probabilities. According to the invention, the problem that maintenance cost and downtime are increased because maintenance personnel can only find a fault source through repeated maintenance attempts because simple sensor alarm or historical data analysis cannot provide a clear traceability path is solved.

Description

Technical Field

[0001] The present application relates to the field of wind power equipment fault analysis, and in particular to a method and device for tracing a single fault of a wind power equipment. Background Art

[0002] As one of the earliest developed and most important renewable energy sources in the integrated energy system, wind energy has great potential in reducing greenhouse gas emissions due to its abundant storage capacity, renewable nature, cleanliness and pollution-free nature. It has become the mainstream direction of global renewable energy development and an important force in promoting industrial upgrading and driving the adjustment of the energy system.

[0003] The existing wind power equipment has poor operating reliability due to the complex and changeable working environment. Downtime due to faults has accounted for 25.6% of the rated power generation time. For wind turbines with an operating life of 20 years, their maintenance costs account for about 20% to 25% of their revenue. The high operation and maintenance costs have become an important factor restricting the development of wind power. There is an urgent need to develop efficient and reliable fault tracing technology.

[0004] At present, common fault tracing methods in wind power equipment usually rely on sensor fault alarms and simple backtracking of historical data. Although these methods can identify the time point when the fault occurs and roughly indicate the fault area, due to the complex operating environment of wind power equipment, these traditional methods are difficult to accurately locate the specific fault source. For example, when there are multiple subsystems in a wind power equipment and a fault occurs, since the operation of each subsystem is often highly coupled, simple sensor alarms or historical data analysis are difficult to determine which subsystem is the initial fault source, resulting in maintenance personnel only being able to find the initial fault source through repeated inspections, thereby increasing maintenance costs and downtime.

[0005] Therefore, there is an urgent need for a single fault tracing method and device for wind power equipment. Summary of the invention

[0006] The present application provides a single fault tracing method and device for wind power equipment, which solves the problem that simple sensor alarms or historical data analysis cannot provide a clear tracing path, resulting in maintenance personnel having to try to find the source of the fault through repeated inspections, thereby increasing maintenance costs and downtime.

[0007] In a first aspect of the present application, a single fault tracing method for wind power equipment is provided, the method comprising: in response to a fault tracing operation for the wind power equipment, obtaining a state monitoring variable corresponding to a target subsystem, the target subsystem being any subsystem in the wind power equipment; calculating the quantity control limit corresponding to the target subsystem through a PCA model; judging whether the wind power equipment is in a fault state according to the quantity control limit; if it is confirmed that the wind power equipment is in a fault state, obtaining a training sample representation, training a SOM network structure and weights through the training sample representation, and judging whether the target subsystem is a faulty subsystem by inputting a state monitoring variable; if it is confirmed that the target subsystem is a faulty subsystem, obtaining a historical fault component corresponding to the target subsystem, and using the state monitoring variable and the historical fault component as Bayesian network nodes to construct a Bayesian network model; using the Bayesian network model, calculating the conditional probabilities corresponding to multiple fault modes of the target subsystem, one fault mode corresponding to one conditional probability; if it is confirmed that the first conditional probability is greater than the second conditional probability, using the fault mode corresponding to the first conditional probability as the fault source corresponding to the wind power equipment, the first conditional probability and the second conditional probability being any two different probabilities among a plurality of different conditional probabilities.

[0008] Optionally, obtain the state monitoring variables corresponding to the target subsystem: obtain the output variables corresponding to the target subsystem, the output variables including the impeller speed variable, the generator power variable and the wind angle variable; use the output variables as the state monitoring variables corresponding to the target subsystem.

[0009] Optionally, the quantity control limit includes a quantity control limit corresponding to the first statistic and a quantity control limit corresponding to the second statistic, and calculating the quantity control limit corresponding to the target subsystem by using the PCA model specifically includes: calculating the first statistic by the following formula:

[0010]

[0011] Among them, SPE is the first statistic, P is the matrix composed of k eigenvectors, I is the unit matrix, x new is a new sample, T is a transposition operation; then the quantity control limit corresponding to the first statistic is calculated according to the following formula:

[0012]

[0013] Among them, Q α is the quantity control limit corresponding to the first statistic, θ r is the eigenvalue corresponding to the sample covariance matrix, and its calculation formula is c α is the confidence limit of the standard normal distribution, h0 is subject to θ r The characteristic statistics are calculated by the control limit of

[0014] The second statistic is calculated by the following formula:

[0015]

[0016] Among them, T 2 is the second statistic, Λ -1 To take the -1 index for each non-zero element in the matrix; then calculate the quantity control limit corresponding to the second statistic according to the following formula:

[0017]

[0018] Among them, T α is the control limit corresponding to the second statistic, 1-α is the confidence level, F α (a,na) ​​is an F distribution with a first degree of freedom and na second degree of freedom.

[0019] Optionally, judging whether the wind power equipment is in a fault state according to the quantity control limit, specifically includes: if it is confirmed that the first statistic is less than the quantity control limit corresponding to the first statistic, and the second statistic is less than the quantity control limit corresponding to the second statistic, judging that the wind power equipment is not in a fault state; if it is confirmed that the first statistic is greater than or equal to the quantity control limit corresponding to the first statistic, or the second statistic is greater than or equal to the quantity control limit corresponding to the second statistic, judging that the wind power equipment is in a fault state.

[0020] Optionally, the SOM network structure and weights are trained by training sample representation, specifically including: S1, obtaining a first sample representation in the training sample representation, where the first sample representation is any sample representation in the training sample representation; S2, obtaining a winning neuron, where the winning neuron is a neuron with the smallest distance from the first sample representation among multiple output neurons; S3, obtaining a connection weight coefficient corresponding to the output neuron in a winning area, and adjusting the connection weight coefficient according to the following formula, where the winning area is the area where the winning neuron is located:

[0021] W ij (t+1)=W ij (t)+η(t)h c,j (t)(X i -W ij (t));

[0022] Among them, W ij (t+1) is the updated weight of the above output neuron, η(t) is the learning rate function, and 0<η(t)<1, the learning rate function is the gain function, h c,j (t) is the domain function; S4, updates the domain function by the following formula:

[0023]

[0024] Among them, d cj is the Euclidean distance between the winning neuron and the activated output neuron in the field, r is the neighborhood radius, and t is the number of iterations; S5, the field radius is updated by the following formula:

[0025]

[0026] Where r(t+1) is the updated domain radius, INT is the rounding function, and T is the total number of training iterations; S6 updates the initial learning rate using the following formula:

[0027]

[0028] Wherein, η(t+1) is the updated learning rate function, and η(0) is the initial learning rate function; S7, let t=t+1, repeat steps S1-S6 until t=T or the domain function converges to complete the training of the SOM network structure and weights.

[0029] Optionally, by inputting state monitoring variables, it is determined whether the target subsystem is a faulty subsystem, specifically including: performing sensitivity analysis on the target subsystem according to the state monitoring variables and based on network slicing, the sensitivity analysis being used to calculate a target sensitivity calculation value corresponding to the target subsystem; determining whether the target sensitivity calculation value is a maximum sensitivity calculation value among multiple sensitivity calculation values, wherein the multiple sensitivity calculation values ​​correspond one-to-one to multiple subsystems; if it is confirmed that the target sensitivity calculation value is the maximum sensitivity calculation value among multiple sensitivity calculation values, then determining that the target subsystem is a faulty subsystem.

[0030] Optionally, calculating the conditional probabilities corresponding to the target subsystem in multiple failure modes respectively includes: calculating the conditional probability by the following formula:

[0031]

[0032] Among them, P(X i |Y) is the conditional probability, X1,…,X n is a complete event group consisting of state monitoring variables and failure modes, and X i ,X j ∈{X1,…,X n}.

[0033] In a second aspect of the present application, a single fault tracing device for wind power equipment is provided, the device comprising a first construction module, a second construction module, a model building module and an analysis and calculation module, wherein:

[0034] The first construction module is used to obtain the state monitoring variables corresponding to the target subsystem in response to the fault tracing operation on the wind power equipment, where the target subsystem is any subsystem in the wind power equipment; and calculate the quantity control limit corresponding to the target subsystem through the PCA model.

[0035] The second construction module is used to determine whether the wind power equipment is in a fault state according to the quantity control limit; if the wind power equipment is confirmed to be in a fault state, the training sample representation is obtained, the SOM network structure and weights are trained through the training sample representation, and the target subsystem is determined to be a faulty subsystem by inputting state monitoring variables.

[0036] The model building module is used to obtain the historical fault components corresponding to the target subsystem if the target subsystem is confirmed to be a faulty subsystem, and use the state monitoring variables and the historical fault components as Bayesian network nodes to build a Bayesian network model.

[0037] The analysis and calculation module is used to calculate the conditional probabilities corresponding to multiple fault modes of the target subsystem through a Bayesian network model, where one fault mode corresponds to one conditional probability; if it is confirmed that the first conditional probability is greater than the second conditional probability, the fault mode corresponding to the first conditional probability is used as the fault source corresponding to the wind power equipment, and the first conditional probability and the second conditional probability are any two different probabilities among multiple different conditional probabilities.

[0038] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any one of the methods described above.

[0039] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to perform any of the above methods.

[0040] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0041] 1. In response to the fault tracing operation for the wind power equipment, the state monitoring variables corresponding to the target subsystem are obtained; the quantity control limit corresponding to the target subsystem is calculated through the PCA model; according to the quantity control limit, whether the wind power equipment is in a fault state is judged; if the wind power equipment is confirmed to be in a fault state, the training sample representation is obtained, the SOM network structure and weight are trained through the training sample representation, and whether the target subsystem is a faulty subsystem is judged by inputting the state monitoring variables; if the target subsystem is confirmed to be a faulty subsystem, the historical fault components corresponding to the target subsystem are obtained, and the state monitoring variables and the historical fault components are used as Bayesian network nodes to construct a Bayesian network model; through the Bayesian network model, the corresponding conditional probabilities of the target subsystem when it is in multiple fault modes are calculated respectively; if it is confirmed that the first conditional probability is greater than the second conditional probability, the fault mode corresponding to the first conditional probability is used as the fault source corresponding to the wind power equipment, and the first conditional probability and the second conditional probability are any two different probabilities among multiple different conditional probabilities, thereby realizing the accurate positioning of the fault source of the wind power equipment and improving the accuracy and efficiency of fault tracing identification.

[0042] 2. By using multiple output variables as the status monitoring variables corresponding to the target subsystem, the operating status of different subsystems can be fully reflected, thereby ensuring that in the subsequent fault tracing operation, the fault source of the wind power equipment can be more comprehensively identified.

[0043] 3. According to the state monitoring variables and based on the network slice, the sensitivity analysis is performed on the target subsystem. The sensitivity analysis is used to calculate the target sensitivity calculation value corresponding to the target subsystem, and to determine whether the target sensitivity calculation value is the maximum sensitivity calculation value among multiple sensitivity calculation values; if it is confirmed that the target sensitivity calculation value is the maximum sensitivity calculation value among multiple sensitivity calculation values, the target subsystem is judged to be a faulty subsystem, thereby improving the accuracy of fault diagnosis and reducing the possibility of misjudgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flow chart of a single fault tracing method for wind power equipment provided in an embodiment of the present application;

[0045] Figure 2 is a line diagram of a first statistic and a second statistic provided in an embodiment of the present application;

[0046] Figure 3 This is a schematic diagram of constructing a Bayesian network model provided in an embodiment of the present application;

[0047] Figure 4 It is a module schematic diagram of a single fault tracing device for wind power equipment provided in an embodiment of the present application;

[0048] Figure 5It is a structural schematic diagram of an electronic device provided in an embodiment of the present application.

[0049] Explanation of the accompanying drawings: 41, first construction module; 42, second construction module; 43, model building module; 44, analysis and calculation module; 501, processor; 502, communication bus; 503, user interface; 504, network interface; 505, memory. DETAILED DESCRIPTION

[0050] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0051] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to and includes any or all possible combinations of one or more listed items.

[0052] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0053] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0054] As one of the earliest developed and most important renewable energy sources in the comprehensive energy system, wind energy has great potential in reducing greenhouse gas emissions due to its abundant storage, renewable, clean and pollution-free characteristics. It has become the mainstream direction of global renewable energy development and an important force in promoting industrial upgrading and energy system adjustment. Driven by the "dual carbon" policy, the wind power industry is expected to become a new opportunity for energy development.

[0055] Data released by the Global Wind Energy Council show that in 2021, the global wind power installed capacity increased by more than 93.6GW, and the cumulative capacity reached 837GW. China's new offshore wind power installed capacity that year was 17GW, accounting for 80% of the global offshore installed capacity increase. According to China's wind power development plan, my country's wind power installed capacity will reach 400GW in 2050, meeting nearly 8.4% of the country's electricity demand. By 2050, this value will be increased to 1TW, meeting 17% of electricity demand. However, the operating reliability of existing wind power equipment is poor due to the complex and changeable working environment. Downtime caused by faults has accounted for 25.6% of the rated power generation time. For wind turbines with an operating life of 20 years, their maintenance costs account for about 20% to 25% of their revenue. The high operation and maintenance costs have become an important factor restricting the development of wind power. There is an urgent need to develop efficient and reliable fault tracing technology.

[0056] At present, common fault tracing methods in wind power equipment usually rely on sensor fault alarms and simple backtracking of historical data. Although these methods can identify the time point when the fault occurs and roughly indicate the fault area, due to the complex operating environment of wind power equipment, these traditional methods are difficult to accurately locate the specific fault source. For example, when there are multiple subsystems in a wind power equipment and a fault occurs, since the operation of each subsystem is often highly coupled, simple sensor alarms or historical data analysis are difficult to determine which subsystem is the initial fault source, resulting in maintenance personnel only being able to find the initial fault source through repeated inspections, thereby increasing maintenance costs and downtime.

[0057] Therefore, there is an urgent need for a single fault tracing method and device for wind power equipment.

[0058] Please refer to Figure 1 , which shows a flow chart of a single fault tracing method for wind power equipment provided in an embodiment of the present application. The method is applied to a server. The flow chart mainly includes the following steps: S101 to S107.

[0059] Step S101 , in response to a fault tracing operation for a wind power device, a state monitoring variable corresponding to a target subsystem is obtained, where the target subsystem is any subsystem in the wind power device.

[0060] Specifically, in an embodiment of the present application, the server can automatically perform fault tracing operations on the wind power equipment in real time, or the user can manually perform fault tracing operations to troubleshoot the fault source in the wind power equipment; the wind power equipment is composed of multiple subsystems. When it is necessary to perform fault tracing operations on the wind power equipment, the state monitoring variables corresponding to each subsystem are obtained. For the convenience of explanation, the present application takes the target subsystem as an example for introduction. The target subsystem can be any subsystem in the wind power equipment, and the subsystems in the wind power equipment include but are not limited to: variable pitch system, gearbox system, frequency conversion system, generator system, yaw system, cabin system, etc. For the acquisition of state monitoring variables, the corresponding output variables of the target subsystem can be used as the state monitoring variables of the target subsystem, and the output variables include but are not limited to: impeller speed variable, generator power variable, wind angle variable, X-axis vibration variable, high-speed shaft drive end bearing temperature variable, machine side power module temperature variable, etc. The output variables corresponding to different subsystems are shown in Table 1.

[0061] [Table 1]

[0062]

[0063] It should be noted that in wind power equipment, there may be subsystems without corresponding output variables. In this case, by integrating it with the associated subsystem, the output variable corresponding to the associated subsystem is used as the output variable corresponding to the subsystem. For example, taking the hydraulic system and the gearbox system as an example, the hydraulic system has no state detection variable, but the fault characteristics of the two systems are the same, so the two can be associated, and the high-speed shaft drive end bearing temperature is used as the state detection variable shared by the two subsystems to perform fault tracing operations. In addition, the purpose of the associated subsystem is to decouple and trace the fault. For subsystems without corresponding output variables, only one subsystem will be found as its associated subsystem, and multiple associations will not be performed.

[0064] Step S102, calculating the quantity control limit corresponding to the target subsystem through the PCA model.

[0065] Specifically, the PCA model can be trained through the output variables corresponding to each subsystem. After the PCA model is completed, the quantity control limit corresponding to the target subsystem is calculated through the PCA model.

[0066] In a possible implementation, step S102 further includes: calculating the first statistic by the following formula:

[0067]

[0068] Among them, SPE is the first statistic, P is the matrix composed of k eigenvectors, I is the unit matrix, x newis a new sample, T is a transposition operation; then the quantity control limit corresponding to the first statistic is calculated according to the following formula: Among them, Q α is the quantity control limit corresponding to the first statistic, θ r is the eigenvalue corresponding to the sample covariance matrix, and its calculation formula is c α is the confidence limit of the standard normal distribution, h0 is subject to θ r The characteristic statistics are calculated by the control limit of

[0069] The second statistic is calculated by the following formula:

[0070]

[0071] Among them, T 2 is the second statistic, Λ -1 To take the -1 index for each non-zero element in the matrix; then calculate the quantity control limit corresponding to the second statistic according to the following formula:

[0072]

[0073] Among them, T α is the control limit corresponding to the second statistic, 1-α is the confidence level, F α (a,na) ​​is an F distribution with a first degree of freedom and na second degree of freedom.

[0074] Specifically, the PCA model is used to determine whether the wind power equipment as a whole is in a fault state, but it is not used to confirm the specific cause of the fault of the wind power equipment. When determining whether the wind power equipment as a whole is in a fault state, it is first necessary to calculate the quantity control limit corresponding to the target subsystem through the PCA model. The quantity control limit includes the quantity control limit corresponding to the first statistic and the quantity control limit corresponding to the second statistic. The calculation steps of the quantity control limit are as follows: multiple output variables are standardized by the following formula, and the standardized processing is used to calculate the standardized value:

[0075]

[0076] Among them, x s.i is the standardized value, x i is the i-th output variable, S i is the standard deviation of multiple output variables; the standardized values ​​are covariance decomposed using the following formula, which is used to calculate the covariance:

[0077]

[0078] Among them, cov(xi ,x i ) is a plurality of output variables and the first statistic is calculated by the following formula:

[0079]

[0080] Among them, SPE is the first statistic, P is a matrix composed of k eigenvectors, and the matrix is ​​[p1, p2, …, p k ], I is the identity matrix, x new is a new sample, T is a transposition operation; then the quantity control limit corresponding to the first statistic is calculated according to the following formula:

[0081]

[0082] Among them, Q α is the quantity control limit corresponding to the first statistic, θ r is the eigenvalue corresponding to the sample covariance matrix, and its calculation formula is c α is the confidence limit of the standard normal distribution, h0 is subject to θ r The characteristic statistics are calculated by the control limit of

[0083] The second statistic is calculated by the following formula:

[0084]

[0085] Among them, T 2 is the second statistic, Λ -1 To take the -1 index for each non-zero element in the matrix; then calculate the quantity control limit corresponding to the second statistic according to the following formula:

[0086]

[0087] Among them, T α is the control limit corresponding to the second statistic, 1-α is the confidence level, F α (a,na) ​​is an F distribution with a first degree of freedom and na second degree of freedom. Figure 2 , which shows a line diagram of a first statistic and a second statistic provided in an embodiment of the present application.

[0088] Step S103: judging whether the wind power equipment is in a fault state according to the quantity control limit.

[0089] Specifically, according to the calculation of the quantity control limit in step S102, if it is confirmed that the first statistic is less than the quantity control limit corresponding to the first statistic, and the second statistic is less than the quantity control limit corresponding to the second statistic, then it is judged that the wind power equipment is not in a fault state; if it is confirmed that the first statistic is greater than or equal to the quantity control limit corresponding to the first statistic, or the second statistic is greater than or equal to the quantity control limit corresponding to the second statistic, then it is judged that the wind power equipment is in a fault state, and the fault tracing operation is continued at this time, and the fault source of the wind power equipment needs to be found.

[0090] Step S104, if it is confirmed that the wind power equipment is in a fault state, a training sample representation is obtained, the SOM network structure and weights are trained through the training sample representation, and by inputting a state monitoring variable, it is determined whether the target subsystem is a faulty subsystem.

[0091] Specifically, when it is confirmed that the wind power equipment is in a fault state, the fault tracing operation is continued, and the fault source of the wind power equipment needs to be found, and the SOM network structure and weights are constructed. The SOM network structure and weights are constructed through the training sample representation. The training sample representation is the corresponding state monitoring variables of each subsystem in the normal operation state when the wind power equipment is in normal operation. After that, by inputting the state monitoring variables into the trained SOM network structure and weights, the SOM network structure and weights are subjected to sensitivity analysis based on network slicing to determine whether the target subsystem is a faulty subsystem.

[0092] In a possible implementation, step S104 further includes: S1, obtaining a first sample representation in the training sample representation, where the first sample representation is any sample representation in the training sample representation; S2, obtaining a winning neuron, where the winning neuron is a neuron with the smallest distance from the first sample representation among multiple output neurons; S3, obtaining a connection weight coefficient corresponding to the output neuron in the winning domain, and adjusting the connection weight coefficient according to the following formula, where the winning domain is the domain where the winning neuron is located: W ij (t+1)=W ij (t)+η(t)h c,j (t)(X i -W ij (t)); where W ij (t+1) is, η(t) is the learning rate function, and 0<η(t)<1, the learning rate function is the gain function, h c,j (t) is the domain function; S4, updates the domain function by the following formula:

[0093]

[0094] Among them, d cjis the Euclidean distance between the winning neuron and the activated output neuron in the field, r is the neighborhood radius, and t is the number of iterations; S5, the field radius is updated by the following formula:

[0095]

[0096] Where r(t+1) is the updated domain radius, INT is the rounding function, and T is the total number of training iterations; S6 updates the initial learning rate using the following formula:

[0097]

[0098] Wherein, η(t+1) is the updated learning rate function, and η(0) is the initial learning rate function; S7, let t=t+1, repeat steps S1-S6 until t=T or the domain function converges to complete the training of the SOM network structure and weights.

[0099] Specifically, a learning process is constructed, and when the learning process completes convergence, the training of the SOM network structure and weights is completed. The training steps of the learning process are as follows: S1, obtain the first sample representation in the training sample representation, and the first sample representation is any sample representation in the training sample representation: When performing this step, first set the initial field, at this time, you can initialize t = 0 and W ij (0), and establish a radius N around the output neuron j. j (0) is the larger initial neighborhood. S2, obtain the winning neuron, which is the neuron with the smallest distance to the first sample representation among multiple output neurons: select any sample X from the state monitoring variable for input, and calculate the distance d between the input sample X and each output neuron j j , and select the output neuron c with the smallest distance as the winning neuron. The calculation formula is:

[0100]

[0101] ‖XW j ‖=min(d j );

[0102] Among them, d j is the distance between the input sample X and each output neuron j, X is the input sample, ‖XW j ‖ is the Euclidean distance, min(d j ) is the minimum distance, W ij (t) is the output neuron weight, X i(t) is the iiith variable in the input sample; S3, obtain the connection weight coefficient corresponding to the output neuron in the winning area, and adjust the connection weight coefficient according to the following formula, the winning area is the area where the winning neuron is located: W ij (t+1)=W ij (t)+η(t)h c,j (t)(X i -W ij (t)); where W ij (t+1) is the updated output neuron weight, η(t) is the learning rate function, and 0<η(t)<1, the learning rate function is the gain function, h c,j (t) is the domain function; S4, updates the domain function by the following formula:

[0103]

[0104] Among them, d cj is the Euclidean distance between the winning neuron and the activated output neuron in the field, r is the neighborhood radius, and t is the number of iterations; S5, the field radius is updated by the following formula:

[0105]

[0106] Where r(t+1) is the updated domain radius, INT is the rounding function, and T is the number of iterations; S6 updates the initial learning rate using the following formula:

[0107]

[0108] Among them, η(t+1) is the updated learning rate function, and η(0) is the initial learning rate function; S7, repeat the above steps S1-S6 until the domain function, domain radius and convergence initial learning rate are all converged to complete the training of the SOM network structure and weights.

[0109] In a possible implementation, step S104 also includes: performing sensitivity analysis on the target subsystem according to the state monitoring variables and based on the network slice, the sensitivity analysis being used to calculate a target sensitivity calculation value corresponding to the target subsystem; judging whether the target sensitivity calculation value is a maximum sensitivity calculation value among multiple sensitivity calculation values, wherein the multiple sensitivity calculation values ​​correspond one-to-one to the multiple subsystems; if it is confirmed that the target sensitivity calculation value is the maximum sensitivity calculation value among the multiple sensitivity calculation values, judging that the target subsystem is a faulty subsystem.

[0110] Specifically, after inputting the state monitoring variable, a sensitivity analysis is performed on each state monitoring variable based on the network slice to obtain the sensitivity calculation value corresponding to each state monitoring variable, where the target subsystem corresponds to the target sensitivity calculation value. If the target subsystem is a faulty subsystem, the target sensitivity calculation value corresponding to the target subsystem should be the maximum calculation value of all sensitivity calculation values. If this condition is met, the state monitoring variable corresponding to the target subsystem is the state monitoring variable that has failed, and the target subsystem corresponding to the state monitoring variable is the subsystem that has failed. The sensitivity calculation formula through network slicing can be calculated as follows:

[0111]

[0112] Among them, ω fe is the sensitivity calculation value, W c After the input state monitoring variable, the weight change at the winning neuron position of the network slice fe, W max is the maximum weight in the network slice fe.

[0113] Step S105: if it is confirmed that the target subsystem is a faulty subsystem, the historical faulty components corresponding to the target subsystem are obtained, and the state monitoring variables and the historical faulty components are used as Bayesian network nodes to construct a Bayesian network model.

[0114] Specifically, if the target subsystem is confirmed to be a faulty subsystem, it is also necessary to obtain the specific fault mode corresponding to the faulty subsystem. The fault modes include but are not limited to: sensor failure, mechanical damage, abnormal temperature, overload operation, component wear, etc. At this time, the historical fault components corresponding to the target subsystem are obtained, and the state monitoring variables and historical fault components are used as Bayesian network nodes to build a Bayesian network model, and the dependencies between the network nodes in the Bayesian network are obtained from the given data set. The given data set includes but is not limited to: historical fault data, sensor record data, equipment operation logs, maintenance records, etc. According to the BN fault path tracing algorithm assumption, the network node is only affected by its parent node, and only the parent node set of the network node participates in Bayesian reasoning, which further simplifies the conditional probability space. By calculating the mutual information between the state monitoring variables as the conditional probability of the node, the calculation formula is as follows:

[0115]

[0116] P(X|Y)=I(x;y);

[0117] Among them, I(x;y) is the mutual information of x and y, x and y are different state monitoring variables, p(x,y) is the joint probability density function of x and y, p(x) is the marginal probability density function of x, and p(y) is the marginal probability density function of y. The conditional probability between nodes is calculated by taking the frequency data of the historical failure modes of the state monitoring variables as follows:

[0118]

[0119] Where P(X|Y) is the conditional probability of the state monitoring variable and the failure mode, n ci-xy is the frequency of failure mode X being observed when a state monitoring variable Y is abnormal, n ci-y The total number of times the condition monitoring variable Y is judged to be abnormal, independent of whether the specific failure mode X occurs.

[0120] Step S106, using the Bayesian network model, calculate the corresponding conditional probabilities when the target subsystem has multiple failure modes, one failure mode corresponds to one conditional probability

[0121] Specifically, the Bayesian network model is used to calculate the corresponding conditional probabilities of the target subsystem in all failure modes.

[0122] Step S107: if it is confirmed that the first conditional probability is greater than the second conditional probability, the fault mode corresponding to the first conditional probability is used as the fault source corresponding to the wind power equipment, and the first conditional probability and the second conditional probability are any two different probabilities among a plurality of different conditional probabilities.

[0123] Specifically, when obtaining the fault mode, the corresponding conditional probabilities are respectively obtained, and the largest conditional probability, that is, the first conditional probability, is found through screening, and the fault mode corresponding to the maximum conditional probability is obtained. At this time, the fault mode is the fault source corresponding to the wind power equipment.

[0124] In a possible implementation, step S107 further includes: calculating a first conditional probability using the following formula:

[0125]

[0126] Among them, P(X i |Y) is the conditional probability, X1,…,X n is a complete event group consisting of state monitoring variables and failure modes, and X i ,X j ∈{X1,…,X n}.

[0127] Specifically, please refer to Figure 3, which shows a schematic diagram of building a Bayesian network model provided by an embodiment of the present application, wherein I1 and I2 are state monitoring variables at the component level, FM1 and FM2 are failure modes at the component level, and the conditional probability between I1 and I2 is calculated based on mutual information. The mutual information of random variables is used to measure the correlation between events.

[0128] Based on the location of the target subsystem and the state monitoring variables detected as faults, the probability of target subsystem failure is set to 1.0 as the prior probability. Taking this as the starting point, Bayesian reasoning is performed using the following formula:

[0129]

[0130] Among them, P(X i |Y) is the conditional probability, X1,…,X n is a complete event group consisting of state monitoring variables and failure modes, and X i ,X j ∈{X1,…,X n By using Bayesian reasoning, the conditional probability of each failure mode occurring in the target subsystem is calculated. The greater the probability, the greater the possibility of the failure mode occurring. Therefore, the failure mode with the largest conditional probability is selected as the failure source.

[0131] The present application adopts the above method to obtain the state monitoring variables corresponding to the target subsystem in response to the fault tracing operation for the wind power equipment; calculate the quantity control limit corresponding to the target subsystem through the PCA model; judge whether the wind power equipment is in a fault state according to the quantity control limit; if it is confirmed that the wind power equipment is in a fault state, obtain the training sample representation, train the SOM network structure and weights through the training sample representation, and judge whether the target subsystem is a faulty subsystem by inputting the state monitoring variables; if it is confirmed that the target subsystem is a faulty subsystem, obtain the historical fault components corresponding to the target subsystem, and use the state monitoring variables and the historical fault components as Bayesian network nodes to construct a Bayesian network model; through the Bayesian network model, calculate the corresponding conditional probabilities when the target subsystem is in multiple fault modes; if it is confirmed that the first conditional probability is greater than the second conditional probability, use the fault mode corresponding to the first conditional probability as the fault source corresponding to the wind power equipment, and the first conditional probability and the second conditional probability are any two different probabilities among multiple different conditional probabilities, thereby achieving accurate positioning of the fault source of the wind power equipment and improving the accuracy and efficiency of fault tracing identification.

[0132] Please refer to Figure 4 , which shows a module schematic diagram of a single fault tracing device for wind power equipment provided in an embodiment of the present application, the device is a server, and the server includes a first construction module 41, a second construction module 42, a model building module 43 and an analysis and calculation module 44, wherein,

[0133] The first construction module 41 is used to obtain the state monitoring variables corresponding to the target subsystem in response to the fault tracing operation on the wind power equipment, where the target subsystem is any subsystem in the wind power equipment; and calculate the quantity control limit corresponding to the target subsystem through the PCA model.

[0134] The second construction module 42 is used to determine whether the wind power equipment is in a fault state according to the quantity control limit; if the wind power equipment is confirmed to be in a fault state, obtain the training sample representation, train the SOM network structure and weights through the training sample representation, and determine whether the target subsystem is a faulty subsystem by inputting the state monitoring variable.

[0135] The model building module 43 is used to obtain the historical fault components corresponding to the target subsystem if it is confirmed that the target subsystem is a faulty subsystem, and use the state monitoring variables and the historical fault components as Bayesian network nodes to build a Bayesian network model.

[0136] The analysis and calculation module 44 is used to calculate the conditional probabilities corresponding to the target subsystem when it is in multiple fault modes through a Bayesian network model, and one fault mode corresponds to one conditional probability; if it is confirmed that the first conditional probability is greater than the second conditional probability, the fault mode corresponding to the first conditional probability is used as the fault source corresponding to the wind power equipment, and the first conditional probability and the second conditional probability are any two different probabilities among multiple different conditional probabilities.

[0137] In one possible implementation, the first construction module 41 is used to obtain the state monitoring variables corresponding to the target subsystem: obtain the output variables corresponding to the target subsystem, the output variables including the impeller speed variable, the generator power variable and the wind angle variable; and use the output variables as the state monitoring variables corresponding to the target subsystem.

[0138] In a possible implementation, the quantity control limit includes a quantity control limit corresponding to the first statistic and a quantity control limit corresponding to the second statistic. The first construction module 41 is used to calculate the quantity control limit corresponding to the target subsystem through the PCA model, specifically including: calculating the first statistic through the following formula:

[0139]

[0140] Among them, SPE is the first statistic, P is the matrix composed of k eigenvectors, I is the unit matrix, x new is a new sample, T is a transposition operation; then the quantity control limit corresponding to the first statistic is calculated according to the following formula:

[0141]

[0142] Among them, Q αis the quantity control limit corresponding to the first statistic, θ r is the eigenvalue corresponding to the sample covariance matrix, and its calculation formula is c α is the confidence limit of the standard normal distribution, h0 is subject to θ r The characteristic statistics are calculated by the control limit of

[0143] The second statistic is calculated by the following formula:

[0144]

[0145] Among them, T 2 is the second statistic, Λ -1 To take the -1 index for each non-zero element in the matrix; then calculate the quantity control limit corresponding to the second statistic according to the following formula:

[0146]

[0147] Among them, T α is the control limit corresponding to the second statistic, 1-α is the confidence level, F α (a,na) ​​is an F distribution with a first degree of freedom and na second degree of freedom.

[0148] In a possible implementation, the second construction module 42 is used to determine whether the wind power equipment is in a fault state based on the quantity control limit, specifically including: if it is confirmed that the first statistic is less than the quantity control limit corresponding to the first statistic, and the second statistic is less than the quantity control limit corresponding to the second statistic, then it is determined that the wind power equipment is not in a fault state; if it is confirmed that the first statistic is greater than or equal to the quantity control limit corresponding to the first statistic, or the second statistic is greater than or equal to the quantity control limit corresponding to the second statistic, then it is determined that the wind power equipment is in a fault state.

[0149] In a possible implementation, the second construction module 42 is used to train the SOM network structure and weights through the training sample representation, specifically including: S1, obtaining the first sample representation in the training sample representation, the first sample representation is any sample representation in the training sample representation; S2, obtaining the winning neuron, the winning neuron is the neuron with the smallest distance from the first sample representation among multiple output neurons; S3, obtaining the connection weight coefficient corresponding to the output neuron in the winning field, and adjusting the connection weight coefficient according to the following formula, the winning field is the field where the winning neuron is located:

[0150] W ij (t+1)=W ij (t)+η(t)h c,j (t)(X i -W ij(t));

[0151] Among them, W ij (t+1) is the updated weight of the above output neuron, η(t) is the learning rate function, and 0<η(t)<1, the learning rate function is the gain function, h c,j (t) is the domain function; S4, updates the domain function by the following formula:

[0152]

[0153] Among them, d cj is the Euclidean distance between the winning neuron and the activated output neuron in the field, r is the neighborhood radius, and t is the number of iterations; S5, the field radius is updated by the following formula:

[0154]

[0155] Where r(t+1) is the updated domain radius, INT is the rounding function, and T is the total number of training iterations; S6 updates the initial learning rate using the following formula:

[0156]

[0157] Wherein, η(t+1) is the updated learning rate function, and η(0) is the initial learning rate function; S7, let t=t+1, repeat steps S1-S6 until t=T or the domain function converges to complete the training of the SOM network structure and weights.

[0158] In a possible implementation, the second construction module 42 is used to determine whether the target subsystem is a faulty subsystem by inputting a state monitoring variable, specifically including: performing a sensitivity analysis on the target subsystem according to the state monitoring variable and based on the network slice, the sensitivity analysis being used to calculate a target sensitivity calculation value corresponding to the target subsystem; determining whether the target sensitivity calculation value is a maximum sensitivity calculation value among multiple sensitivity calculation values, wherein the multiple sensitivity calculation values ​​correspond one-to-one to multiple subsystems; if it is confirmed that the target sensitivity calculation value is the maximum sensitivity calculation value among multiple sensitivity calculation values, determining that the target subsystem is a faulty subsystem.

[0159] In a possible implementation, the analysis and calculation module 44 is used to calculate the conditional probabilities corresponding to the target subsystem in multiple failure modes, specifically including: calculating the conditional probability by the following formula:

[0160]

[0161] Among them, P(X i |Y) is the conditional probability, X1,…,X nis a complete event group consisting of state monitoring variables and failure modes, and X i ,X j ∈{X1,…,X n}.

[0162] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0163] The present application also provides an electronic device. Figure 5 , Figure 5 1 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include: at least one processor 501, at least one communication bus 502, a user interface 503, at least one network interface 504, and a memory 505.

[0164] The communication bus 502 is used to realize the connection and communication between these components.

[0165] The user interface 503 may include a display screen (Display) and a camera (Camera), and the optional user interface 503 may also include a standard wired interface and a wireless interface.

[0166] The network interface 504 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0167] Among them, the processor 501 may include one or more processing cores. The processor 501 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 505, and calling data stored in the memory 505. Optionally, the processor 501 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 501 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 501, and it can be implemented separately through a chip.

[0168] Among them, the memory 505 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 505 includes a non-transitory computer-readable storage medium. The memory 505 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 505 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 505 may also be optionally at least one storage device located away from the aforementioned processor 501. Refer to Figure 3 , the memory 505 as a computer storage medium may include an operating system, a network communication module, a user interface module and a single fault tracing application program for wind power equipment.

[0169] exist Figure 3In the electronic device shown, the user interface 503 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 501 can be used to call the single fault tracing application for wind power equipment stored in the memory 505. When executed by one or more processors 501, the electronic device executes one or more of the methods described in the above embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited to the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0170] The present application also provides a computer-readable storage medium, which stores instructions. When executed by one or more processors, the electronic device executes one or more of the methods described in the above embodiments.

[0171] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0172] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0173] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0174] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0175] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0176] The above are only exemplary embodiments disclosed in this application, and the scope of the disclosure of this application cannot be limited by this. That is, any equivalent changes and modifications made according to the teachings disclosed in this application are still within the scope of the disclosure of this application. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other implementation schemes disclosed in this application.

[0177] This application is intended to cover any modifications, uses or adaptive changes disclosed in this application, which follow the general principles disclosed in this application and include common knowledge or customary technical means in the technical field not recorded in this application.

Claims

1. A single fault tracing method for wind power equipment, characterized in that: The method comprises: In response to a fault tracing operation for the wind power equipment, a state monitoring variable corresponding to a target subsystem is acquired, wherein the target subsystem is any subsystem in the wind power equipment; Calculating the quantity control limit corresponding to the target subsystem through the PCA model; determining whether the wind power equipment is in a fault state according to the quantity control limit; If it is confirmed that the wind power equipment is in the fault state, obtaining a training sample representation, training a SOM network structure and weights through the training sample representation, and determining whether the target subsystem is a faulty subsystem by inputting the state monitoring variable; If it is confirmed that the target subsystem is the faulty subsystem, a historical fault component corresponding to the target subsystem is obtained, and the state monitoring variables and the historical fault component are used as Bayesian network nodes to construct a Bayesian network model; By using the Bayesian network model, the conditional probabilities corresponding to the target subsystem being in multiple failure modes are calculated, and one failure mode corresponds to one conditional probability; If it is confirmed that the first conditional probability is greater than the second conditional probability, the fault mode corresponding to the first conditional probability is used as the fault source corresponding to the wind power equipment, and the first conditional probability and the second conditional probability are any two different probabilities among the multiple different conditional probabilities.

2. The method according to claim 1, characterized in that: The state monitoring variables corresponding to the target subsystem are obtained: Obtaining output variables corresponding to the target subsystem, wherein the output variables include an impeller speed variable, a generator power variable, and a wind angle variable; The output variable is used as the state monitoring variable corresponding to the target subsystem.

3. The method according to claim 1, characterized in that The quantity control limit includes a quantity control limit corresponding to the first statistic and a quantity control limit corresponding to the second statistic. The quantity control limit corresponding to the target subsystem is calculated by the PCA model, specifically including: The first statistic is calculated by the following formula: Where SPE is the first statistic, P is a matrix consisting of k eigenvectors, I is the unit matrix, x new is a new sample, T is a transposition operation; then the quantity control limit corresponding to the first statistic is calculated according to the following formula: Among them, Q α is the quantity control limit corresponding to the first statistic, θ r is the eigenvalue corresponding to the sample covariance matrix, and its calculation formula is c α is the confidence limit of the standard normal distribution, h0 is subject to θ r The characteristic statistics are calculated by the control limit of The second statistic is calculated by the following formula: Among them, T 2 is the second statistic, Λ -1 is to take the -1 index for each non-zero element in the matrix; then the quantity control limit corresponding to the second statistic is calculated according to the following formula: Among them, T α is the control limit corresponding to the second statistic, 1-α is the confidence level, F α (a,na) ​​is an F distribution with a first degree of freedom and na second degree of freedom.

4. The method according to claim 3, characterized in that: The determining, according to the quantity control limit, whether the wind power equipment is in a fault state specifically includes: If it is confirmed that the first statistic is smaller than the quantity control limit corresponding to the first statistic, and the second statistic is smaller than the quantity control limit corresponding to the second statistic, it is determined that the wind power equipment is not in the fault state; If it is confirmed that the first statistic is greater than or equal to the quantity control limit corresponding to the first statistic, or the second statistic is greater than or equal to the quantity control limit corresponding to the second statistic, it is determined that the wind power equipment is in the fault state.

5. The method according to claim 1, characterized in that The training of the SOM network structure and weights by characterizing the training samples specifically includes: S1, obtaining a first sample representation in the training sample representation, where the first sample representation is any sample representation in the training sample representation; S2, obtaining a winning neuron, where the winning neuron is a neuron with the smallest distance to the first sample representation among multiple output neurons; S3, obtaining the connection weight coefficient corresponding to the output neuron in the winning area, and adjusting the connection weight coefficient according to the following formula, wherein the winning area is the area where the winning neuron is located: W ij (t+1)=W ij (t)+η(t)h c,j (t)(X i -W ij (t)); Among them, W ij (t+1) is the updated weight of the output neuron, η(t) is the learning rate function, and 0<η(t)<1, the learning rate function is the gain function, h c,j (t) is the domain function; S4, updating the domain function by the following formula: Among them, d cj is the Euclidean distance between the winning neuron and the output neuron activated in the field, r is the neighborhood radius, and t is the number of iterations; S5, update the field radius using the following formula: Among them, r(t+1) is the updated domain radius, INT is the rounding function, and T is the total number of training iterations; S6, updating the initial learning rate by the following formula: Wherein, η(t+1) is the updated learning rate function, and η(0) is the initial learning rate function; S7, let t=t+1, repeat steps S1-S6 until t=T or the domain function converges to complete the training of the SOM network structure and weights.

6. The method according to claim 1, characterized in that The step of judging whether the target subsystem is a faulty subsystem by inputting the state monitoring variable specifically includes: Performing a sensitivity analysis on the target subsystem according to the state monitoring variable and based on the network slice, wherein the sensitivity analysis is used to calculate a target sensitivity calculation value corresponding to the target subsystem; Determining whether the target sensitivity calculation value is a maximum sensitivity calculation value among a plurality of sensitivity calculation values, wherein the plurality of sensitivity calculation values ​​correspond one-to-one to the plurality of subsystems; If it is confirmed that the target sensitivity calculation value is the maximum sensitivity calculation value among the multiple sensitivity calculation values, the target subsystem is determined to be the faulty subsystem.

7. The method according to claim 1, characterized in that The calculating of the conditional probabilities corresponding to the target subsystem in multiple failure modes respectively specifically includes: The conditional probability is calculated by the following formula: Among them, P(X i |Y) is the conditional probability, X1,…,X n is a complete event group consisting of the state monitoring variables and the failure modes, and X i ,X j ∈{X1,…,X n }.

8. A single fault tracing device for wind power equipment, characterized in that: The device comprises a first construction module, a second construction module, a model building module and an analysis and calculation module, wherein: The first construction module is used to obtain a state monitoring variable corresponding to a target subsystem in response to a fault tracing operation on the wind power equipment, wherein the target subsystem is any subsystem in the wind power equipment; obtain a PCA model, and calculate a quantity control limit corresponding to the target subsystem through the PCA model; The second construction module is used to determine whether the wind power equipment is in a fault state according to the quantity control limit; if it is confirmed that the wind power equipment is in the fault state, obtain a training sample representation, train the SOM network structure and weights through the training sample representation, and determine whether the target subsystem is a faulty subsystem by inputting the state monitoring variable; The model building module is used to obtain the historical fault component corresponding to the target subsystem if it is confirmed that the target subsystem is the fault subsystem, and use the state monitoring variables and the historical fault component as Bayesian network nodes to build a Bayesian network model; The analysis and calculation module is used to calculate the conditional probabilities corresponding to the target subsystem when it is in multiple fault modes through the Bayesian network model, and one fault mode corresponds to one conditional probability; if it is confirmed that the first conditional probability is greater than the second conditional probability, the fault mode corresponding to the first conditional probability is used as the fault source corresponding to the wind power equipment, and the first conditional probability and the second conditional probability are any two different probabilities among the multiple different conditional probabilities.

9. An electronic device, characterized in that: It includes a processor, a communication bus, a user interface, a network interface and a memory, wherein the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is performed.

Images