Reliability analysis and maintenance method based on wind power generation multi-component degradation modeling

Abstract

The invention provides a reliability analysis and maintenance method based on wind power generation multi-component degradation modeling, which relates to the field of wind power generation and comprises the following steps: establishing a blade degradation evaluation model, a transmission mechanism degradation evaluation model and a generator degradation evaluation model according to operating environment characteristics, respectively determining current blade degradation information, current transmission mechanism degradation information and current generator degradation information; future blade degradation information, future transmission mechanism degradation information and future generator degradation information are determined through a joint degradation prediction model according to the current blade degradation information, the current transmission mechanism degradation information, the current generator degradation information and the operation state information of the wind generating set; according to the current blade degradation information, the current transmission mechanism degradation information, the current generator degradation information, the future blade degradation information, the future transmission mechanism degradation information and the future generator degradation information, the maintenance scheme is determined, and the method has the advantages that fault pre-judgment is achieved, and the maintenance efficiency is improved.

Description

Technical Field

[0001] The invention relates to the field of wind power generation, and in particular to a reliability analysis and maintenance method based on degradation modeling of multiple components of wind power generation. Background Art

[0002] Wind energy is a renewable energy source with great potential and has received widespread attention from countries around the world. Wind turbines are the core components of wind power generation systems, which realize the conversion of mechanical energy into electrical energy. Wind turbines are mostly installed in remote and harsh environments such as the sea, plateaus, and mountainous areas. The operating conditions are extremely harsh, which makes the performance of wind turbine components easily degraded.

[0003] The equipment protection strategy commonly adopted by existing wind turbines is over-limit protection. Although this equipment protection strategy has practical significance, it also has obvious defects. The control method and control strategy based on over-limit protection consider fewer factors for the status of wind turbines and have a relatively simple processing method. In fact, the real status of wind turbines has changed a long time before the over-limit shutdown, and is in a sub-healthy or abnormal state. In addition, the maintenance of wind turbines mostly relies on the practical experience of operation and maintenance personnel. After a new fault occurs, they can only rely on checking the fault handling manual and wind turbine drawings to find the fault location, analyze the cause of the fault, and finally give a fault repair plan. The whole process requires operation and maintenance personnel to gradually judge based on past experience. Not only is the maintenance efficiency low, but the work tasks for operation and maintenance personnel are arduous, consuming a lot of manpower and material resources, and the professional level of operation and maintenance personnel is high. For new operation and maintenance personnel, they can only gradually accumulate experience through existing technical information and the types of faults that can be encountered. The improvement speed is slow, and it is difficult to meet the needs of wind turbine maintenance work with diverse faults and heavy maintenance tasks.

[0004] Therefore, it is necessary to provide a reliability analysis and maintenance method based on degradation modeling of multiple wind power generation components, which can be used to perform reliability analysis of multiple wind power generation components, thereby realizing fault prediction and improving maintenance efficiency. Summary of the invention

[0005] The present invention provides a reliability analysis and maintenance method based on degradation modeling of multiple components of wind power generation, comprising: acquiring operating environment characteristics of a wind generator set; establishing a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model according to the operating environment characteristics; establishing a joint degradation prediction model according to the operating environment characteristics; collecting operating status information of the wind generator set; determining current blade degradation information, current transmission mechanism degradation information and current generator degradation information respectively according to the operating status information of the wind generator set through the blade degradation assessment model, the transmission mechanism degradation assessment model and the generator degradation assessment model; determining future blade degradation information, future transmission mechanism degradation information and future generator degradation information according to the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information and the operating status information of the wind generator set through the joint degradation prediction model; and determining a maintenance plan according to the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information and the future blade degradation information, the future transmission mechanism degradation information and the future generator degradation information.

[0006] Furthermore, according to the operating environment characteristics, a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model are established, including: determining multiple blade degradation performance factors, transmission mechanism degradation performance factors and generator degradation performance factors; acquiring multiple sets of simulation data according to the multiple blade degradation performance factors, transmission mechanism degradation performance factors and generator degradation performance factors, wherein the multiple sets of simulation data include performance characteristic information of simulated wind turbine generator sets in different degradation states under multiple simulated operating environments; determining a first target blade degradation performance factor, a first target transmission mechanism degradation performance factor and a first target generator degradation performance factor corresponding to the operating environment characteristics according to the multiple sets of simulation data; establishing a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model according to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics.

[0007] Further, according to the multiple sets of simulation data, the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics are determined, including: for each of the simulated operating environments, according to the multiple sets of simulation data corresponding to the simulated operating environment, the first correlation coefficient between each of the blade degradation performance factors and the blade degradation state, the first correlation coefficient between each of the transmission mechanism degradation performance factors and the transmission mechanism degradation state, and the first correlation coefficient between each of the generator degradation performance factors and the generator degradation state are calculated; according to the first correlation coefficient between each of the blade degradation performance factors and the blade degradation state, the first correlation coefficient between each of the transmission mechanism degradation performance factors and the transmission mechanism degradation state, and the first correlation coefficient between each of the generator degradation performance factors and the generator degradation state, the first blade degradation performance factor, the first transmission mechanism degradation performance factor and the first generator degradation performance factor corresponding to the simulated operating environment are screened; according to the multiple sets of simulation data corresponding to the simulated operating environment, the first correlation coefficient between any two first blade degradation performance factors is calculated. second correlation coefficient, second correlation coefficients between any two first transmission mechanism degradation performance factors and the transmission mechanism degradation state, and second correlation coefficients between any two of the generator degradation performance factors and the generator degradation state; determine the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor, and the first target generator degradation performance factor corresponding to the simulated operating environment according to the second correlation coefficients of any two first blade degradation performance factors, the second correlation coefficients between any two first transmission mechanism degradation performance factors and the transmission mechanism degradation state, and the second correlation coefficients between any two of the generator degradation performance factors and the generator degradation state; calculate the environmental similarity between the operating environment characteristics and each of the simulated operating environments, and determine the target simulated operating environment according to the environmental similarity; determine the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor, and the first target generator degradation performance factor corresponding to the operating environment characteristics according to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor, and the first target generator degradation performance factor corresponding to the target simulated operating environment.

[0008] Further, determining a first correlation coefficient between each of the leaf degradation performance factors and the leaf degradation state includes: For each blade degradation state, the time series simulation data corresponding to each blade degradation performance factor is determined according to the simulation data corresponding to the blade degradation state, and variational modal decomposition is performed on the time series simulation data corresponding to the blade degradation performance factor to extract the variational modal decomposition features corresponding to the blade degradation performance factor; for each blade degradation performance factor, the first correlation coefficient between the blade degradation performance factor and the blade degradation state is calculated according to the variational modal decomposition features corresponding to the blade degradation performance factor under different blade degradation states.

[0009] Furthermore, according to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics, a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model are established, including: according to the first correlation coefficient between each of the blade degradation performance factors and the degradation state, the first correlation coefficient between each of the transmission mechanism degradation performance factors and the degradation state and the first correlation coefficient between each of the generator degradation performance factors and the degradation state under each simulated operating environment, determining the key blade degradation performance factor, the key transmission mechanism degradation performance factor and the key generator degradation performance factor; according to the key blade degradation performance factor, the key transmission mechanism degradation performance factor and the key generator degradation performance factor, establishing a basic blade degradation assessment model, a basic transmission mechanism degradation assessment model and a basic generator degradation assessment model; for any two of the simulated operating environments, calculating the environmental similarity of the two simulated operating environments, and according to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the two simulated operating environments. The target generator degradation performance factor calculates the performance factor similarity; clusters the multiple simulated operating environments according to the environmental similarity and performance factor similarity of any two of the simulated operating environments to determine multiple environmental clustering clusters; determines a first target environment clustering cluster from the multiple environmental clustering clusters according to the operating environment characteristics; establishes a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model according to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics and the basic blade degradation assessment model, the basic transmission mechanism degradation assessment model and the basic generator degradation assessment model; generates multiple training samples through a sample generation model according to multiple groups of simulation data corresponding to the multiple simulated operating environments included in the first target environment clustering cluster and the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics; trains the blade degradation assessment model, the transmission mechanism degradation assessment model and the generator degradation assessment model through the multiple training samples.

[0010] Furthermore, according to the operating environment characteristics, a joint degradation prediction model is established, including: for each of the simulated operating environments, according to multiple groups of simulation data corresponding to the simulated operating environment, calculating the third correlation coefficient between any first target blade degradation performance factor and any first target transmission mechanism degradation performance factor, the third correlation coefficient between any first target generator degradation performance factor and any first target transmission mechanism degradation performance factor, and the third correlation coefficient between any first target generator degradation performance factor and any first target blade degradation performance factor corresponding to the simulated operating environment; and according to the third correlation coefficient between any first target blade degradation performance factor and any first target transmission mechanism degradation performance factor, the third correlation coefficient between any first target generator degradation performance factor and any first target transmission mechanism degradation performance factor, The third correlation coefficient and the third correlation coefficient between any first target generator degradation performance factor and any first target blade degradation performance factor are used to select the second blade degradation performance factor, the second transmission mechanism degradation performance factor and the second generator degradation performance factor from the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the simulated operating environment; according to the second blade degradation performance factor, the second transmission mechanism degradation performance factor and the second generator degradation performance factor corresponding to the target simulated operating environment, the second blade degradation performance factor, the second transmission mechanism degradation performance factor and the second generator degradation performance factor corresponding to the operating environment characteristics are determined; according to the second blade degradation performance factor, the second transmission mechanism degradation performance factor and the second generator degradation performance factor corresponding to the operating environment characteristics, a joint degradation prediction model is established.

[0011] Furthermore, the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information are respectively determined according to the operating status information of the wind turbine generator set through the blade degradation assessment model, the transmission mechanism degradation assessment model and the generator degradation assessment model, including: extracting a first blade state feature vector, a first transmission mechanism state feature vector and a first generator state feature vector from the operating status information of the wind turbine generator set according to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics; determining the current blade degradation information according to the first blade state feature vector through the blade degradation assessment model; determining the current transmission mechanism degradation information according to the first transmission mechanism state feature vector through the transmission mechanism degradation assessment model; and determining the current generator degradation information according to the first generator state feature vector through the generator degradation assessment model.

[0012] Furthermore, the future blade degradation information, future transmission mechanism degradation information and future generator degradation information are determined by the joint degradation prediction model according to the current blade degradation information, current transmission mechanism degradation information and current generator degradation information and the operating status information of the wind turbine generator set, including: extracting the second blade state feature vector, the second transmission mechanism state feature vector and the second generator state feature vector from the operating status information of the wind turbine generator set according to the second blade degradation performance factor, the second transmission mechanism degradation performance factor and the second generator degradation performance factor corresponding to the operating environment characteristics; and determining the future blade degradation information, future transmission mechanism degradation information and future generator degradation information according to the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information and the second blade state feature vector, the second transmission mechanism state feature vector and the second generator state feature vector through the joint degradation prediction model.

[0013] Further, according to the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information and the future blade degradation information, the future transmission mechanism degradation information and the future generator degradation information, a maintenance plan is determined, including: S11, establishing a maintenance knowledge map, wherein the maintenance knowledge map is used to record the relationship between degradation characteristics and maintenance methods; S12, according to the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information, determining whether to perform maintenance, if so, executing S13, if not, executing S14; S13, according to the The maintenance plan is determined based on the knowledge graph, the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information; S14, according to the future blade degradation information, the future transmission mechanism degradation information and the future generator degradation information, determine whether to perform maintenance, if so, execute S15, if not, after updating the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information, execute S12; S15, according to the knowledge graph, the future blade degradation information, the future transmission mechanism degradation information and the future generator degradation information, determine the maintenance plan.

[0014] Furthermore, a maintenance plan is determined based on the knowledge graph, the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information, including: determining a maintenance process based on the knowledge graph, the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information; obtaining the historical power generation information of the wind turbine generator set and the maintenance process to determine the optimal maintenance time period; obtaining the background information of the maintenance personnel and the maintenance process to determine the optimal maintenance personnel, wherein the maintenance plan at least includes a maintenance process, an optimal maintenance time period and an optimal maintenance personnel.

[0015] Compared with the prior art, the reliability analysis and maintenance method based on wind power generation multi-component degradation modeling provided by the present invention has at least the following beneficial effects: 1. This method can accurately predict the degradation of each component by establishing a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model. This helps to detect potential faults in time and avoid downtime caused by sudden faults, thereby improving the reliability of the entire wind turbine. Through the joint degradation prediction model, the degradation of multiple components and their mutual influence can be comprehensively considered, so as to more accurately predict the degradation trend of each component in the future. This helps to formulate a more reasonable maintenance plan and reduce unnecessary maintenance costs. This method can determine the maintenance plan based on the current and future degradation information of each component. This helps maintenance personnel to choose appropriate maintenance methods such as preventive maintenance and post-failure maintenance according to actual conditions, thereby optimizing the maintenance strategy. Through accurate independent prediction and joint degradation prediction, maintenance time and resources can be reasonably arranged to avoid excessive or insufficient maintenance. This helps to reduce maintenance costs and improve economic benefits. By timely detecting and predicting faults, maintenance measures can be taken quickly to reduce downtime. This helps to improve the operating efficiency of wind turbines and increase power generation.

[0016] 2. By determining the degradation performance factors of multiple blades, transmission mechanisms, and generators, and obtaining simulation data based on these factors, the degradation of each component under different operating environments can be more accurately reflected. This helps to accurately identify the key factors that affect the degradation of each component, thereby improving the accuracy and reliability of the model. By calculating the correlation coefficient and screening the target degradation performance factors, redundant information can be excluded and the factors that have the greatest impact on the degradation state can be retained. This helps to optimize the parameter settings of the model and improve the prediction accuracy and generalization ability of the model.

[0017] 3. By determining the key blade degradation performance factors, the key transmission mechanism degradation performance factors and the key generator degradation performance factors, and according to the key blade degradation performance factors, the key transmission mechanism degradation performance factors and the key generator degradation performance factors, the basic blade degradation assessment model, the basic transmission mechanism degradation assessment model and the basic generator degradation assessment model are established, which provides a data basis for the subsequent establishment of blade degradation assessment models, transmission mechanism degradation assessment models and generator degradation assessment models according to the operating environment characteristics, and improves the efficiency of establishing blade degradation assessment models, transmission mechanism degradation assessment models and generator degradation assessment models according to the operating environment characteristics. In addition, a sample generation model is used to generate multiple training samples according to the simulated operating environment and key degradation performance factors in the first target environment cluster. This can ensure the diversity and representativeness of the training samples and help the model learn more comprehensive degradation characteristics. Using the generated training samples to train the degradation assessment model can ensure the accuracy and reliability of the model in the real operating environment. At the same time, through continuous iteration and optimization, the prediction accuracy and adaptability of the model can be further improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein: Figure 1 It is a flowchart of a reliability analysis and maintenance method based on degradation modeling of wind power generation multiple components according to some embodiments of this specification; Figure 2 It is a schematic diagram of a process of establishing a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model according to some embodiments of this specification; Figure 3 It is a schematic diagram of the knowledge graph shown in some embodiments of this specification. DETAILED DESCRIPTION

[0019] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0020] Figure 1 is a flow chart of a reliability analysis and maintenance method based on degradation modeling of wind power generation multiple components according to some embodiments of this specification, such as Figure 1 As shown, the reliability analysis and maintenance method based on wind power generation multi-component degradation modeling may include the following process.

[0021] Step 110, obtaining operating environment characteristics of the wind turbine generator set.

[0022] Specifically, the operating environment characteristics of the wind turbine generator set can be the characteristics of the environment in which the wind turbine generator set is located, which have a certain impact on the degradation of the wind turbine generator set. For example, the instability of wind speed and wind direction will affect the output power and stability of the wind turbine generator set. Too low or too high wind speed may cause the unit to have a reduced operating efficiency, while changes in wind direction may cause additional wear of mechanical parts. Extreme temperature conditions will have an adverse effect on the materials and components of the wind turbine generator set. High temperature may cause the material to soften and reduce efficiency, while low temperature may cause the material to become brittle and fragile. Strong wind, high temperature, low temperature, humidity, salt spray, etc., these factors will have a common impact on components such as blades, transmission mechanisms and generators, causing their performance to gradually degrade. For example, high temperature and humid environment will accelerate the oxidation and deterioration of lubricating oil, resulting in increased wear of the transmission mechanism and generator; salt spray environment will cause corrosion and rust of components such as blades and transmission mechanisms.

[0023] Step 120: establishing a blade degradation assessment model, a transmission mechanism degradation assessment model, and a generator degradation assessment model according to the operating environment characteristics.

[0024] In some embodiments, step 120 specifically includes: Determine a plurality of blade degradation performance factors, transmission mechanism degradation performance factors and generator degradation performance factors, for example, the plurality of blade degradation performance factors may include a blade vibration factor, a blade stress factor, a blade surface color factor and a blade surface roughness factor, the plurality of transmission mechanism degradation performance factors may include a transmission mechanism vibration factor, a lubricating oil temperature factor, etc., the plurality of generator degradation performance factors may include a generator current factor, a generator voltage factor, a generator power factor factor, a generator insulation resistance factor, a generator vibration factor, a generator noise factor, etc.; According to a plurality of blade degradation performance factors, a transmission mechanism degradation performance factor and a generator degradation performance factor, a plurality of sets of simulation data are obtained, wherein the plurality of sets of simulation data include performance characteristic information of simulated wind turbine generator sets in different degradation states under a plurality of simulated operating environments; Determine, according to the plurality of sets of simulation data, a first target blade degradation performance factor, a first target transmission mechanism degradation performance factor, and a first target generator degradation performance factor corresponding to the operating environment characteristics; According to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics, a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model are established.

[0025] Specifically, when the blades are operating normally, their vibration amplitude and frequency should be within a certain range. If the vibration is abnormal, such as increased amplitude or frequency change, it may mean that the blades have problems such as wear, cracks or imbalance. The stress distribution of the blades is monitored by stress sensors. If uneven stress distribution is found, especially in key areas such as the root or leading edge of the blades, it may mean that the blades have problems such as material aging or external impact. The surface of the blades may be eroded by natural environments such as wind, sand, rain and snow, resulting in increased surface roughness and coating shedding. In addition, the blades may also be affected by ultraviolet radiation, resulting in material aging and discoloration. These problems will be reflected through changes in surface conditions.

[0026] When the transmission mechanism is running, if the vibration and noise increase significantly, it may mean that the gears, bearings and other parts are worn, loose or damaged. The lubricating oil in the transmission mechanism should maintain a certain temperature range during normal operation. If the oil temperature is too high, it may mean that the transmission mechanism is overheated, poorly lubricated or oil contaminated.

[0027] When the generator is operating normally, its current and voltage should remain stable. If there are fluctuations or abnormal changes, it may mean that the generator has problems such as short circuit, open circuit or excitation system failure. Power factor is an important indicator to measure the quality of the power output of the generator. If the power factor decreases, it may mean that the generator is overloaded, has power loss or insufficient reactive power compensation. An insulation resistance test can detect the insulation resistance value between the generator winding and the ground. If the insulation resistance decreases, it may mean that the winding is aging, damp or contaminated. Similar to the transmission mechanism, if the generator has abnormal vibration and noise during operation, it may also mean that there are degradation problems such as looseness, wear or damage inside.

[0028] In some embodiments, determining a first target blade degradation performance factor, a first target transmission mechanism degradation performance factor, and a first target generator degradation performance factor corresponding to the operating environment characteristics according to multiple sets of simulation data includes: For each simulated operating environment, according to multiple groups of simulation data corresponding to the simulated operating environment, the first correlation coefficient between each blade degradation performance factor and the blade degradation state, the first correlation coefficient between each transmission mechanism degradation performance factor and the transmission mechanism degradation state, and the first correlation coefficient between each generator degradation performance factor and the generator degradation state are calculated; according to the first correlation coefficient between each blade degradation performance factor and the blade degradation state, the first correlation coefficient between each transmission mechanism degradation performance factor and the transmission mechanism degradation state, and the first correlation coefficient between each generator degradation performance factor and the generator degradation state, the first blade degradation performance factor, the first transmission mechanism degradation performance factor, and the first generator degradation performance factor corresponding to the simulated operating environment are screened; According to a plurality of groups of simulation data corresponding to the simulated operating environment, second correlation coefficients of any two first blade degradation performance factors, second correlation coefficients of any two first transmission mechanism degradation performance factors and transmission mechanism degradation states, and second correlation coefficients of any two generator degradation performance factors and generator degradation states are calculated; and according to the second correlation coefficients of any two first blade degradation performance factors, second correlation coefficients of any two first transmission mechanism degradation performance factors and transmission mechanism degradation states, and second correlation coefficients of any two generator degradation performance factors and generator degradation states, a first target blade degradation performance factor, a first target transmission mechanism degradation performance factor, and a first target generator degradation performance factor corresponding to the simulated operating environment are determined; Calculate the environmental similarity between the operating environment characteristics and each simulated operating environment, and determine the target simulated operating environment according to the environmental similarity. For example, the operating environment characteristics and the characteristics of the simulated operating environment are represented as vectors, and the cosine similarity or Euclidean distance between the two vectors is calculated to measure the environmental similarity between the operating environment characteristics and the simulated operating environment, and the simulated operating environment whose environmental similarity is greater than the environmental similarity threshold is used as the target simulated operating environment; According to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the target simulated operating environment, the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics are determined. For example, the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the target simulated operating environment are used as the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics.

[0029] In some embodiments, for a blade degradation performance factor having a time series characteristic (for example, a blade vibration factor, a blade stress factor, etc.), determining a first correlation coefficient between the blade degradation performance factor and the blade degradation state includes: For each blade degradation state, determine the time series simulation data corresponding to each blade degradation performance factor according to the simulation data corresponding to the blade degradation state, perform variational modal decomposition on the time series simulation data corresponding to the blade degradation performance factor, and extract the variational modal decomposition features corresponding to the blade degradation performance factor (for example, the center frequency, bandwidth, amplitude, etc. of the sub-signal obtained after variational modal decomposition); For each blade degradation performance factor, a first correlation coefficient between the blade degradation performance factor and the blade degradation state is calculated according to the variational mode decomposition characteristics corresponding to the blade degradation performance factor under different blade degradation states.

[0030] For example, for a blade vibration factor, the time series simulation data corresponding to the blade vibration factor may include vibration data at multiple simulation time points.

[0031] For leaf degradation performance factors with time series characteristics, the first correlation coefficient between the leaf degradation performance factor and the leaf degradation state can be calculated according to the following formula: , , , , , , , in, is the first correlation coefficient between the i-th leaf degradation performance factor and the leaf degradation state, , and is the weight, , is the central frequency correlation coefficient between the i-th leaf degradation performance factor and the leaf degradation state, is the bandwidth correlation coefficient between the i-th blade degradation performance factor and the blade degradation state, is the amplitude correlation coefficient between the i-th leaf degradation performance factor and the leaf degradation state, is the correlation coefficient between the i-th leaf degradation performance factor and the k-th sub-signal center frequency of the leaf degradation state, is the weight, , K is the total number of sub-signals obtained after variational mode decomposition, is the center frequency of the kth sub-signal obtained after variational mode decomposition under the nth degradation state corresponding to the ith blade degradation performance factor, is the total number of degenerate states, is the state value corresponding to the nth degradation state, is the weight, , is the correlation coefficient between the i-th blade degradation performance factor and the k-th sub-signal bandwidth of the blade degradation state, is the bandwidth of the kth sub-signal obtained after variational mode decomposition under the nth degradation state corresponding to the ith blade degradation performance factor, is the weight, , is the correlation coefficient between the i-th leaf degradation performance factor and the k-th sub-signal amplitude of the leaf degradation state, is the amplitude of the kth sub-signal obtained after variational mode decomposition under the nth degradation state corresponding to the ith blade degradation performance factor.

[0032] For leaf degradation performance factors without temporal characteristics (e.g., leaf surface color factor and leaf surface roughness factor), the first correlation coefficient between the leaf degradation performance factor and the leaf degradation state can be calculated according to the following formula: , in, is the first correlation coefficient between the jth leaf degradation performance factor and the leaf degradation state, It is the factor value in the nth degradation state corresponding to the jth blade degradation performance factor. For example, for the blade surface color factor, the factor value in the nth degradation state corresponding to the blade surface color factor may be the average of the RGB values ​​of the pixels calculated based on the collected blade surface image. For another example, for the blade surface roughness factor, the factor value in the nth degradation state corresponding to the blade surface roughness factor may be the blade surface roughness calculated based on the collected blade surface image.

[0033] The method of determining the first correlation coefficient between each transmission mechanism degradation performance factor and the transmission mechanism degradation state and the first correlation coefficient between each generator degradation performance factor and the generator degradation state is similar to the method of determining the first correlation coefficient between each blade degradation performance factor and the blade degradation state, and will not be repeated here.

[0034] For any two first leaf degradation performance factors with time series characteristics, the second correlation coefficient of any two first leaf degradation performance factors can be calculated according to the following formula: , , , , , , , in, is the second correlation coefficient between the i-th first leaf degradation performance factor and the j-th first leaf degradation performance factor, , and is the weight, , is the central frequency correlation coefficient between the i-th first leaf degradation performance factor and the j-th first leaf degradation performance factor, is the bandwidth correlation coefficient between the i-th first blade degradation performance factor and the j-th first blade degradation performance factor, is the amplitude correlation coefficient between the i-th first leaf degradation performance factor and the j-th first leaf degradation performance factor, is the correlation coefficient between the kth sub-signal center frequency of the i-th first blade degradation performance factor and the j-th first blade degradation performance factor, is the weight, , K is the total number of sub-signals obtained after variational mode decomposition. When the number of sub-signals corresponding to the i-th first leaf degradation performance factor is inconsistent with the number of sub-signals corresponding to the j-th first leaf degradation performance factor, the smaller value of the number of sub-signals corresponding to the i-th first leaf degradation performance factor and the number of sub-signals corresponding to the j-th first leaf degradation performance factor is taken. is the center frequency of the kth sub-signal obtained after performing variational mode decomposition under the nth degradation state corresponding to the jth first blade degradation performance factor, is the weight, , is the bandwidth correlation coefficient of the kth sub-signal of the i-th first leaf degradation performance factor and the j-th first leaf degradation performance factor, is the bandwidth of the kth sub-signal obtained after variational mode decomposition under the nth degradation state corresponding to the jth first blade degradation performance factor, is the weight, , is the amplitude correlation coefficient of the kth sub-signal of the i-th first leaf degradation performance factor and the j-th first leaf degradation performance factor, is the amplitude of the kth sub-signal obtained after variational mode decomposition under the nth degradation state corresponding to the jth first blade degradation performance factor.

[0035] For any two first blade degradation performance factors, one with a timing feature and the other without a timing feature, the method of calculating the second correlation coefficient of the two first blade degradation performance factors is similar to the method of calculating the first correlation coefficient between the blade degradation performance factor without a timing feature and the blade degradation state, which will not be repeated here. Among them, the factor value in the nth degradation state corresponding to the blade vibration factor can be the vibration frequency, vibration amplitude, etc., and the factor value in the nth degradation state corresponding to the blade stress factor can be the stress value, etc.

[0036] For any two first blade degradation performance factors without timing characteristics, the method of calculating the second correlation coefficient of the two first blade degradation performance factors is similar to the method of calculating the first correlation coefficient between the blade degradation performance factor without timing characteristics and the blade degradation state, which will not be repeated here.

[0037] The method of calculating the second correlation coefficient between any two first transmission mechanism degradation performance factors and the transmission mechanism degradation state and the second correlation coefficient between any two generator degradation performance factors and the generator degradation state is similar to the method of calculating the second correlation coefficient of any two first blade degradation performance factors, and will not be repeated here.

[0038] Figure 2 is a flow chart of establishing a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model according to some embodiments of this specification, such as Figure 2 As shown, in some embodiments, according to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics, a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model are established, including: Determine the key blade degradation performance factor, the key transmission mechanism degradation performance factor and the key generator degradation performance factor according to the first correlation coefficient between each blade degradation performance factor and the degradation state, the first correlation coefficient between each transmission mechanism degradation performance factor and the degradation state and the first correlation coefficient between each generator degradation performance factor and the degradation state under each simulated operating environment; According to the key blade degradation performance factors, the key transmission mechanism degradation performance factors and the key generator degradation performance factors, a basic blade degradation assessment model, a basic transmission mechanism degradation assessment model and a basic generator degradation assessment model are established, wherein the basic blade degradation assessment model, the basic transmission mechanism degradation assessment model and the basic generator degradation assessment model may be a Long Short-Term Memory (LSTM) network model; For any two simulated operating environments, the environmental similarity of the two simulated operating environments is calculated, and the performance factor similarity is calculated according to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the two simulated operating environments. For example, the overlap of the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the two simulated operating environments can be calculated as the performance factor similarity; According to the environmental similarity and performance factor similarity of any two simulated operating environments, a plurality of simulated operating environments are clustered to determine a plurality of environmental clusters; Determine a first target environment cluster from a plurality of environment clusters according to the operating environment characteristics; Establishing a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model according to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics and the basic blade degradation assessment model, the basic transmission mechanism degradation assessment model and the basic generator degradation assessment model; Generate multiple training samples through a sample generation model according to multiple groups of simulation data corresponding to multiple simulated operating environments included in the first target environment cluster and first target blade degradation performance factors, first target transmission mechanism degradation performance factors and first target generator degradation performance factors corresponding to the operating environment characteristics, wherein the sample generation model can be a variational autoencoder, a Generative Adversarial Networks (GAN) model, etc.; The blade degradation assessment model, the transmission mechanism degradation assessment model and the generator degradation assessment model are trained through multiple training samples.

[0039] Specifically, the key value corresponding to the blade degradation performance factor may be calculated based on the following formula according to the first correlation coefficient between the blade degradation performance factor and the degradation state under each simulated operating environment: , in, is the key value corresponding to the i-th leaf degradation performance factor, is the first correlation coefficient between the degradation performance factor of the i-th blade and the degradation state under the e-th simulated operating environment, is the total number of simulated running environments.

[0040] The leaf degradation performance factor with a critical value greater than the critical value threshold is taken as the critical leaf degradation performance factor.

[0041] The method of determining the key transmission mechanism degradation performance factor and the key generator degradation performance factor is similar to the method of determining the key blade degradation performance factor, which will not be repeated here.

[0042] The input of the basic blade degradation assessment model corresponds to the key blade degradation performance factor, the basic transmission mechanism degradation assessment model corresponds to the key transmission mechanism degradation performance factor, and the basic generator degradation assessment model corresponds to the key generator degradation performance factor.

[0043] For any two simulated operating environments, the environmental similarity and performance factor similarity of the two simulated operating environments can be weighted and summed to obtain the comprehensive similarity of the two simulated operating environments. A clustering algorithm (for example, a K-means clustering algorithm, etc.) is used to cluster multiple simulated operating environments according to the comprehensive similarity of any two simulated operating environments to determine multiple environmental clustering clusters.

[0044] According to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics, the basic blade degradation assessment model, the basic transmission mechanism degradation assessment model and the basic generator degradation assessment model are fine-tuned to establish the blade degradation assessment model, the transmission mechanism degradation assessment model and the generator degradation assessment model.

[0045] Step 130: Establish a joint degradation prediction model according to the operating environment characteristics.

[0046] In some embodiments, step 130 includes: For each simulated operating environment, according to a plurality of groups of simulation data corresponding to the simulated operating environment, calculate the third correlation coefficient between any first target blade degradation performance factor and any first target transmission mechanism degradation performance factor, the third correlation coefficient between any first target generator degradation performance factor and any first target transmission mechanism degradation performance factor, and the third correlation coefficient between any first target generator degradation performance factor and any first target blade degradation performance factor corresponding to the simulated operating environment; based on the third correlation coefficient between any first target blade degradation performance factor and any first target transmission mechanism degradation performance factor, the third correlation coefficient between any first target generator degradation performance factor and any first target transmission mechanism degradation performance factor, and the third correlation coefficient between any first target generator degradation performance factor and any first target blade degradation performance factor, select the second blade degradation performance factor, the second transmission mechanism degradation performance factor, and the second generator degradation performance factor from the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor, and the first target generator degradation performance factor corresponding to the simulated operating environment; Determine the second blade degradation performance factor, the second transmission mechanism degradation performance factor, and the second generator degradation performance factor corresponding to the operating environment characteristics according to the second blade degradation performance factor, the second transmission mechanism degradation performance factor, and the second generator degradation performance factor corresponding to the target simulated operating environment; A joint degradation prediction model is established according to the second blade degradation performance factor, the second transmission mechanism degradation performance factor and the second generator degradation performance factor corresponding to the operating environment characteristics, wherein the joint degradation prediction model can be a Long Short-Term Memory (LSTM) model.

[0047] Specifically, the method of calculating the third correlation coefficient between any first target blade degradation performance factor and any first target transmission mechanism degradation performance factor corresponding to the simulated operating environment, the third correlation coefficient between any first target generator degradation performance factor and any first target transmission mechanism degradation performance factor, and the third correlation coefficient between any first target generator degradation performance factor and any first target blade degradation performance factor is similar to the method of calculating the second correlation coefficient of any two first blade degradation performance factors, the second correlation coefficient between any two first transmission mechanism degradation performance factors and the transmission mechanism degradation state, and the second correlation coefficient between any two generator degradation performance factors and the generator degradation state, and will not be repeated here.

[0048] The second blade degradation performance factor, the second transmission mechanism degradation performance factor and the second generator degradation performance factor can be screened according to the following process: S31. For each first target blade degradation performance factor, according to the third correlation coefficient between the first target blade degradation performance factor and any first target transmission mechanism degradation performance factor and the third correlation coefficient between the first target blade degradation performance factor and any first target generator degradation performance factor, calculate the mean value of the third correlation coefficient corresponding to the first target blade degradation performance factor; S32. For each first target transmission mechanism degradation performance factor, according to the third correlation coefficient between the first target transmission mechanism degradation performance factor and any first target blade degradation performance factor and the third correlation coefficient between the first target transmission mechanism degradation performance factor and any first target generator degradation performance factor, calculate the mean value of the third correlation coefficient corresponding to the first target transmission mechanism degradation performance factor; S33. For each first target generator degradation performance factor, according to the third correlation coefficient between the first target generator degradation performance factor and any first target blade degradation performance factor and the third correlation coefficient between the first target generator degradation performance factor and any first target transmission mechanism degradation performance factor, calculate the mean value of the third correlation coefficient corresponding to the first target generator degradation performance factor; S34, determining whether the third correlation coefficient mean corresponding to each first target blade degradation performance factor, the third correlation coefficient mean corresponding to each first target transmission mechanism degradation performance factor, and the third correlation coefficient mean corresponding to each first target generator degradation performance factor are all greater than or equal to the third correlation coefficient mean threshold, if so, completing the screening, if not, executing S35; S35. Filter out the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor whose third correlation coefficient mean is less than the third correlation coefficient mean threshold, and execute S31 for the remaining first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor.

[0049] Step 140: Collect the operating status information of the wind turbine generator set.

[0050] Specifically, the operating status information of the wind turbine generator set may include: 1. Basic operating parameters: Wind speed and wind direction: directly affect the output power of wind turbines.

[0051] Speed ​​and power: reflect the current operating status and efficiency of the generator.

[0052] Temperature parameters: including ambient temperature, generator temperature, transmission mechanism temperature, etc., used to monitor whether the equipment is overheating.

[0053] 2. Blade operation status information: Blade vibration factor: By monitoring the vibration of the blades in the wind, it is possible to determine whether the blades have structural damage or imbalance.

[0054] Blade stress factor: The stress condition of the blade under wind load is monitored by stress sensors to evaluate the strength and durability of the blade.

[0055] Blade surface color factor: Color changes may indicate aging, corrosion or contamination of the blade surface.

[0056] Blade surface roughness factor: Increased roughness may mean that the blade surface is being worn or eroded.

[0057] 3. Transmission mechanism operating status information: Transmission mechanism vibration factor: Abnormal vibration may indicate wear or failure of components such as gears and bearings.

[0058] Lubricant Temperature Factor: Excessive temperature may indicate poor lubrication or excessive friction.

[0059] Lubricant quality factor: By detecting changes in the composition of the lubricant, its contamination level and replacement timing can be assessed.

[0060] 4. Generator operating status information: Generator Current Factor: Current fluctuations may indicate a problem with the electrical connections or a fault within the generator.

[0061] Generator voltage factor: Voltage instability may be caused by generator winding problems or grid faults.

[0062] Generator Power Factor Factor: Power factor deviations from normal values ​​may indicate reduced generator efficiency or unbalanced grid load.

[0063] Generator Insulation Resistance Factor: A drop in insulation resistance may mean there is a risk of leakage in the generator windings.

[0064] Generator Vibration Factor: Abnormal vibration may indicate a mechanical fault inside the generator.

[0065] Generator Noise Factor: Increased noise may indicate that the generator is not running smoothly or there is friction.

[0066] Step 150, according to the operation status information of the wind turbine generator set, current blade degradation information, current transmission mechanism degradation information and current generator degradation information are determined respectively through the blade degradation assessment model, the transmission mechanism degradation assessment model and the generator degradation assessment model.

[0067] In some embodiments, step 150 includes: Extracting a first blade state feature vector, a first transmission mechanism state feature vector, and a first generator state feature vector from the operating state information of the wind turbine generator set according to a first target blade degradation performance factor, a first target transmission mechanism degradation performance factor, and a first target generator degradation performance factor corresponding to the operating environment characteristics; Determining current blade degradation information according to the first blade state feature vector through a blade degradation assessment model; Determining current transmission mechanism degradation information according to the first transmission mechanism state characteristic vector through the transmission mechanism degradation assessment model; The current generator degradation information is determined according to the first generator state characteristic vector through the generator degradation assessment model.

[0068] Step 160, determining future blade degradation information, future transmission mechanism degradation information and future generator degradation information according to current blade degradation information, current transmission mechanism degradation information and current generator degradation information and wind turbine generator operating status information through a joint degradation prediction model.

[0069] In some embodiments, step 160 includes: Extracting a second blade state feature vector, a second transmission mechanism state feature vector and a second generator state feature vector from the operating state information of the wind turbine generator set according to the second blade degradation performance factor, the second transmission mechanism degradation performance factor and the second generator degradation performance factor corresponding to the operating environment characteristics; The future blade degradation information, future transmission mechanism degradation information and future generator degradation information are determined by a joint degradation prediction model based on the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information and the second blade state eigenvector, the second transmission mechanism state eigenvector and the second generator state eigenvector.

[0070] Step 170, determining a maintenance plan based on current blade degradation information, current transmission mechanism degradation information, current generator degradation information, and future blade degradation information, future transmission mechanism degradation information, and future generator degradation information.

[0071] In some embodiments, step 170 includes: S11. Establishing a maintenance knowledge graph, wherein the maintenance knowledge graph is used to record the relationship between degradation characteristics and maintenance methods; S12, judging whether to perform maintenance according to the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information, if yes, executing S13, if no, executing S14; S13, determining a maintenance plan according to the knowledge graph, current blade degradation information, current transmission mechanism degradation information, and current generator degradation information; S14, judging whether to perform maintenance according to the future blade degradation information, the future transmission mechanism degradation information and the future generator degradation information, if yes, executing S15, if not, updating the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information, and then executing S12; S15. Determine a maintenance plan based on the knowledge graph, future blade degradation information, future transmission mechanism degradation information, and future generator degradation information.

[0072] Specifically, Figure 3 It is a schematic diagram of the knowledge graph shown in some embodiments of this specification. As shown in the figure, the maintenance knowledge graph records the relationship between the degradation characteristics of different components and the maintenance methods.

[0073] In some embodiments, a maintenance plan is determined based on the knowledge graph, the current blade degradation information, the current transmission mechanism degradation information, and the current generator degradation information, including: Determine the maintenance process according to the knowledge graph, current blade degradation information, current transmission mechanism degradation information and current generator degradation information. For example, the maintenance process may be determined according to the knowledge graph, current blade degradation information, current transmission mechanism degradation information and current generator degradation information through a solution generation model, wherein the solution generation model may be a convolutional neural network (CNN) model; Obtain historical power generation information and maintenance procedures of wind turbines to determine the optimal maintenance time period. For example, the time required for maintenance can be predicted based on the maintenance procedure, and based on historical power generation information, a time period with a duration that meets the maintenance time predicted by the maintenance procedure and with less power generation can be selected as the optimal maintenance time period; Obtain the background information and maintenance process of the maintenance personnel and determine the optimal maintenance personnel, wherein the maintenance plan at least includes the maintenance process, the optimal maintenance time period and the optimal maintenance personnel. For example, the skills required for maintenance can be determined based on the maintenance process, and based on the background information of the maintenance personnel, the maintenance personnel who meet the skills required for maintenance determined by the maintenance process can be selected as the optimal maintenance personnel.

[0074] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A reliability analysis and maintenance method based on multi-component degradation modeling of wind power generation, characterized in that: include: Obtain the operating environment characteristics of wind turbines; According to the operating environment characteristics, a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model are established; Establishing a joint degradation prediction model according to the operating environment characteristics; Collecting the operating status information of the wind turbine generator set; Determining the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information respectively according to the operating status information of the wind turbine generator set through the blade degradation assessment model, the transmission mechanism degradation assessment model and the generator degradation assessment model; Determining future blade degradation information, future transmission mechanism degradation information and future generator degradation information according to the current blade degradation information, current transmission mechanism degradation information and current generator degradation information and the operating state information of the wind turbine generator set through the joint degradation prediction model; A maintenance plan is determined according to the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information and the future blade degradation information, the future transmission mechanism degradation information and the future generator degradation information.

2. The reliability analysis and maintenance method based on wind power generation multi-component degradation modeling according to claim 1 is characterized in that: According to the operating environment characteristics, a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model are established, including: Determine a plurality of blade degradation performance factors, a transmission mechanism degradation performance factor and a generator degradation performance factor; Acquire multiple sets of simulation data according to the multiple blade degradation performance factors, transmission mechanism degradation performance factors and generator degradation performance factors, wherein the multiple sets of simulation data include performance characteristic information of simulated wind turbine generator sets in different degradation states under multiple simulated operating environments; Determining, according to the plurality of sets of simulation data, a first target blade degradation performance factor, a first target transmission mechanism degradation performance factor, and a first target generator degradation performance factor corresponding to the operating environment characteristics; According to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics, a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model are established.

3. The reliability analysis and maintenance method based on wind power generation multi-component degradation modeling according to claim 2 is characterized in that: Determining, according to the plurality of sets of simulation data, a first target blade degradation performance factor, a first target transmission mechanism degradation performance factor, and a first target generator degradation performance factor corresponding to the operating environment characteristics, including: For each of the simulated operating environments, according to a plurality of sets of simulation data corresponding to the simulated operating environments, a first correlation coefficient between each of the blade degradation performance factors and the blade degradation state, a first correlation coefficient between each of the transmission mechanism degradation performance factors and the transmission mechanism degradation state, and a first correlation coefficient between each of the generator degradation performance factors and the generator degradation state are calculated; according to the first correlation coefficient between each of the blade degradation performance factors and the blade degradation state, the first correlation coefficient between each of the transmission mechanism degradation performance factors and the transmission mechanism degradation state, and the first correlation coefficient between each of the generator degradation performance factors and the generator degradation state, the first blade degradation performance factor, the first transmission mechanism degradation performance factor, and the first generator degradation performance factor corresponding to the simulated operating environment are screened. performance factor, calculating, according to a plurality of groups of simulation data corresponding to the simulated operating environment, second correlation coefficients of any two first blade degradation performance factors, second correlation coefficients of any two first transmission mechanism degradation performance factors and transmission mechanism degradation states, and second correlation coefficients of any two generator degradation performance factors and generator degradation states, and determining, according to the second correlation coefficients of any two first blade degradation performance factors, second correlation coefficients of any two first transmission mechanism degradation performance factors and transmission mechanism degradation states, and second correlation coefficients of any two generator degradation performance factors and generator degradation states, a first target blade degradation performance factor, a first target transmission mechanism degradation performance factor, and a first target generator degradation performance factor corresponding to the simulated operating environment; Calculating the environmental similarity between the operating environment characteristics and each of the simulated operating environments, and determining a target simulated operating environment according to the environmental similarity; According to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the target simulated operating environment, the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics are determined.

4. The reliability analysis and maintenance method based on wind power generation multi-component degradation modeling according to claim 3 is characterized in that: Determining a first correlation coefficient between each of the leaf degradation performance factors and the leaf degradation state includes: For each blade degradation state, determine the time series simulation data corresponding to each blade degradation performance factor according to the simulation data corresponding to the blade degradation state, perform variational modal decomposition on the time series simulation data corresponding to the blade degradation performance factor, and extract the variational modal decomposition features corresponding to the blade degradation performance factor; For each of the blade degradation performance factors, a first correlation coefficient between the blade degradation performance factor and the blade degradation state is calculated according to variational mode decomposition characteristics corresponding to the blade degradation performance factor under different blade degradation states.

5. The reliability analysis and maintenance method based on wind power generation multi-component degradation modeling according to claim 3 is characterized in that: According to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics, a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model are established, including: Determine the key blade degradation performance factor, the key transmission mechanism degradation performance factor and the key generator degradation performance factor according to the first correlation coefficient between each blade degradation performance factor and the degradation state, the first correlation coefficient between each transmission mechanism degradation performance factor and the degradation state and the first correlation coefficient between each generator degradation performance factor and the degradation state under each simulated operating environment; According to the key blade degradation performance factor, the key transmission mechanism degradation performance factor and the key generator degradation performance factor, a basic blade degradation assessment model, a basic transmission mechanism degradation assessment model and a basic generator degradation assessment model are established; For any two of the simulated operating environments, calculating the environmental similarity of the two simulated operating environments, and calculating the performance factor similarity according to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the two simulated operating environments; Clustering the multiple simulated operating environments according to the environmental similarity and performance factor similarity of any two of the simulated operating environments to determine multiple environmental clusters; Determine a first target environment cluster from the plurality of environment clusters according to the operating environment characteristics; Establishing a blade degradation assessment model, a transmission mechanism degradation assessment model and a generator degradation assessment model according to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor and the first target generator degradation performance factor corresponding to the operating environment characteristics and the basic blade degradation assessment model, the basic transmission mechanism degradation assessment model and the basic generator degradation assessment model; Generate multiple training samples through a sample generation model according to multiple groups of simulation data corresponding to multiple simulated operating environments included in the first target environment cluster and first target blade degradation performance factors, first target transmission mechanism degradation performance factors and first target generator degradation performance factors corresponding to the operating environment characteristics; The blade degradation assessment model, the transmission mechanism degradation assessment model and the generator degradation assessment model are trained through the multiple training samples.

6. The reliability analysis and maintenance method based on wind power generation multi-component degradation modeling according to claim 3 is characterized in that: According to the operating environment characteristics, a joint degradation prediction model is established, including: For each of the simulated operating environments, according to a plurality of groups of simulation data corresponding to the simulated operating environments, calculate a third correlation coefficient between any first target blade degradation performance factor and any first target transmission mechanism degradation performance factor, a third correlation coefficient between any first target generator degradation performance factor and any first target transmission mechanism degradation performance factor, and a third correlation coefficient between any first target generator degradation performance factor and any first target blade degradation performance factor corresponding to the simulated operating environment; and select a second blade degradation performance factor, a second transmission mechanism degradation performance factor, and a second generator degradation performance factor from the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor, and the first target generator degradation performance factor corresponding to the simulated operating environment according to the third correlation coefficient between any first target blade degradation performance factor and any first target transmission mechanism degradation performance factor, the third correlation coefficient between any first target generator degradation performance factor and any first target transmission mechanism degradation performance factor, and the third correlation coefficient between any first target generator degradation performance factor and any first target blade degradation performance factor; Determine the second blade degradation performance factor, the second transmission mechanism degradation performance factor, and the second generator degradation performance factor corresponding to the operating environment characteristics according to the second blade degradation performance factor, the second transmission mechanism degradation performance factor, and the second generator degradation performance factor corresponding to the target simulated operating environment; A joint degradation prediction model is established according to the second blade degradation performance factor, the second transmission mechanism degradation performance factor and the second generator degradation performance factor corresponding to the operating environment characteristics.

7. The reliability analysis and maintenance method based on wind power generation multi-component degradation modeling according to any one of claims 2 to 6, characterized in that: The blade degradation assessment model, the transmission mechanism degradation assessment model and the generator degradation assessment model are used to respectively determine the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information according to the operating status information of the wind turbine generator set, including: Extracting a first blade state feature vector, a first transmission mechanism state feature vector, and a first generator state feature vector from the operating state information of the wind turbine generator set according to the first target blade degradation performance factor, the first target transmission mechanism degradation performance factor, and the first target generator degradation performance factor corresponding to the operating environment characteristics; Determining the current blade degradation information according to the first blade state feature vector by using the blade degradation assessment model; Determining the current transmission mechanism degradation information according to the first transmission mechanism state characteristic vector by using the transmission mechanism degradation assessment model; The current generator degradation information is determined according to the first generator state characteristic vector through the generator degradation assessment model.

8. The reliability analysis and maintenance method based on wind power generation multi-component degradation modeling according to claim 6 is characterized in that: Determining future blade degradation information, future transmission mechanism degradation information and future generator degradation information according to the current blade degradation information, current transmission mechanism degradation information and current generator degradation information and the operating state information of the wind turbine generator set through the joint degradation prediction model, including: Extracting a second blade state feature vector, a second transmission mechanism state feature vector and a second generator state feature vector from the operating state information of the wind turbine generator set according to the second blade degradation performance factor, the second transmission mechanism degradation performance factor and the second generator degradation performance factor corresponding to the operating environment characteristics; The future blade degradation information, future transmission mechanism degradation information and future generator degradation information are determined through the joint degradation prediction model according to the current blade degradation information, current transmission mechanism degradation information and current generator degradation information and the second blade state characteristic vector, the second transmission mechanism state characteristic vector and the second generator state characteristic vector.

9. The reliability analysis and maintenance method based on wind power generation multi-component degradation modeling according to any one of claims 1 to 6, characterized in that: Determining a maintenance plan according to the current blade degradation information, the current transmission mechanism degradation information, and the current generator degradation information and the future blade degradation information, the future transmission mechanism degradation information, and the future generator degradation information, including: S11. Establishing a maintenance knowledge graph, wherein the maintenance knowledge graph is used to record the relationship between degradation characteristics and maintenance methods; S12, judging whether to perform maintenance according to the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information, if yes, executing S13, if no, executing S14; S13, determining a maintenance plan according to the knowledge graph, the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information; S14, judging whether to perform maintenance according to the future blade degradation information, the future transmission mechanism degradation information and the future generator degradation information, if yes, executing S15, if no, updating the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information, and then executing S12; S15. Determine a maintenance plan based on the knowledge graph, the future blade degradation information, the future transmission mechanism degradation information, and the future generator degradation information.

10. The reliability analysis and maintenance method based on wind power generation multi-component degradation modeling according to claim 9 is characterized in that: Determining a maintenance plan according to the knowledge graph, the current blade degradation information, the current transmission mechanism degradation information, and the current generator degradation information, including: Determine a maintenance process according to the knowledge graph, the current blade degradation information, the current transmission mechanism degradation information and the current generator degradation information; Acquire historical power generation information of the wind turbine generator set and the maintenance process to determine an optimal maintenance time period; The background information of the maintenance personnel and the maintenance process are obtained to determine the optimal maintenance personnel, wherein the maintenance plan at least includes the maintenance process, the optimal maintenance time period and the optimal maintenance personnel.

Images