CN117231439A - Fault diagnosis and control method and device for wind turbine generator set and electronic equipment - Google Patents

Abstract

The application provides a fault diagnosis and control method, a device and electronic equipment of a wind turbine, wherein the method comprises the following steps: acquiring state parameters and environment parameters of a wind turbine to be detected; acquiring the fault type and the fault grade of the wind turbine to be detected according to the state parameters and the environment parameters; selecting a target control scheme from the candidate control schemes according to the fault type and the fault level; according to the target control scheme, the target control parameters and the target control strategy of the wind turbine to be detected are determined, the accuracy and the high efficiency of fault diagnosis of the wind turbine are improved, the corresponding target control scheme can be selected according to different fault types and fault grades, the control parameters and the strategy of the wind turbine can be dynamically adjusted, the target control parameters and the target control strategy of the wind turbine to be detected are determined, the running efficiency and the reliability of the wind turbine are optimized, the abrasion and the loss of the wind turbine are reduced, and the service life of the wind turbine is prolonged.

Description

Fault diagnosis and control method and device for wind turbine generator set and electronic equipment

Technical Field

The application relates to the technical field of wind power generation, in particular to a fault diagnosis and control method and device for a wind turbine generator, and electronic equipment.

Background

At present, a wind turbine generator fault diagnosis scheme is often based on a quasi-static state parameter detection method, for example: the method is characterized in that vibration signals, oil parameters, temperature parameters, current, voltage and other parameters of an output end of a generator are detected at certain moments to perform fault diagnosis, however, the requirements on wind turbine faults are not met, a wind turbine fault diagnosis scheme is often analyzed based on a single or a small number of signal sources, correlation and complementarity between various state parameters and environment information of the wind turbine cannot be fully utilized, the running state and fault characteristics of the wind turbine cannot be comprehensively reflected, uncertainty and complexity in the running process of the wind turbine are not considered when the wind turbine fault diagnosis scheme is judged based on experience or a model, the variable working conditions and fault types of the wind turbine cannot be met, the wind turbine fault diagnosis scheme reduces the accuracy and the practicability of performing fault diagnosis on the wind turbine, influences the running efficiency and the reliability of the wind turbine, and how to perform fault diagnosis and control on the wind turbine so as to improve the accuracy of fault diagnosis of the wind turbine and the running efficiency of the wind turbine.

Disclosure of Invention

The object of the present application is to solve at least to some extent one of the technical problems in the art described above.

The first aspect of the application provides a fault diagnosis and control method of a wind turbine generator, comprising the following steps: acquiring state parameters and environment parameters of a wind turbine to be detected; acquiring the fault type and the fault grade of the wind turbine to be detected according to the state parameters and the environment parameters; selecting a target control scheme from the candidate control schemes according to the fault type and the fault level; and determining target control parameters and target control strategies of the wind turbine to be detected according to the target control scheme.

The second aspect of the application provides a fault diagnosis and control device of a wind turbine generator, comprising: the first acquisition module is used for acquiring state parameters and environment parameters of the wind turbine to be detected; the second acquisition module is used for acquiring the fault type and the fault grade of the wind turbine to be detected according to the state parameters and the environment parameters; the selecting module is used for selecting a target control scheme from the candidate control schemes according to the fault type and the fault level; and the determining module is used for determining target control parameters and target control strategies of the wind turbine to be detected according to the target control scheme.

An embodiment of a third aspect of the present application provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; the memory stores instructions executable by the at least one processor, so that the at least one processor can execute the fault diagnosis and control method of the wind turbine generator set provided in the first aspect of the present application.

An embodiment of a fourth aspect of the present application provides a non-transitory computer readable storage medium storing computer instructions for causing a computer to execute the method for diagnosing and controlling a fault of a wind turbine provided in the first aspect of the present application.

An embodiment of a fifth aspect of the present application provides a computer program product, which when executed by an instruction processor in the computer program product, performs the method for diagnosing and controlling a fault of a wind turbine generator provided in the first aspect of the present application.

According to the fault diagnosis and control method and device for the wind turbine, the state parameters and the environment parameters of the wind turbine to be detected are obtained, the fault type and the fault grade of the wind turbine to be detected are obtained according to the state parameters and the environment parameters, the target control scheme is selected from the candidate control schemes according to the fault type and the fault grade, and the target control parameters and the target control strategy of the wind turbine to be detected are determined according to the target control scheme.

Additional aspects and advantages of the application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the application.

Drawings

The foregoing and/or additional aspects and advantages of the application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:

FIG. 1 is a flow chart of a method for diagnosing and controlling faults of a wind turbine according to an embodiment of the present application;

FIG. 2 is a flow chart of a method for diagnosing and controlling faults of a wind turbine according to another embodiment of the present application;

FIG. 3 is a flow chart of a method for diagnosing and controlling faults of a wind turbine according to another embodiment of the present application;

FIG. 4 is a flowchart of a method for diagnosing and controlling faults of a wind turbine according to another embodiment of the present application;

FIG. 5 is a flow chart of a method for diagnosing and controlling faults of a wind turbine according to another embodiment of the present application;

FIG. 6 is a flow chart of a method for diagnosing and controlling faults of a wind turbine according to another embodiment of the present application;

FIG. 7 is a flow chart of a method for diagnosing and controlling faults of a wind turbine according to another embodiment of the present application;

FIG. 8 is a schematic structural diagram of a fault diagnosis and control apparatus for a wind turbine according to an embodiment of the present application;

Fig. 9 is a block diagram of an electronic device in accordance with an embodiment of the application.

Detailed Description

Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative and intended to explain the present application and should not be construed as limiting the application.

The following describes a fault diagnosis and control method, a device, electronic equipment and a medium of a wind turbine generator set according to an embodiment of the application with reference to the accompanying drawings.

FIG. 1 is a schematic flow chart of a method for diagnosing and controlling faults of a wind turbine according to an embodiment of the present application, as shown in FIG. 1, the method includes:

s101, acquiring state parameters and environment parameters of the wind turbine to be detected.

It should be noted that, the specific mode of acquiring the state parameter and the environmental parameter of the wind turbine to be detected is not limited, and the method can be selected according to actual conditions.

Optionally, an internal sensor of the wind turbine may be used to obtain a state parameter of the wind turbine to be detected.

For example, state parameters such as temperature, pressure, power and rotation speed of the wind turbine to be detected can be obtained through internal sensors such as temperature sensors, pressure sensors, power sensors and rotation speed sensors of the wind turbine.

Optionally, an external sensor of the wind turbine may be used to obtain an environmental parameter of the wind turbine to be detected.

For example, the environmental parameters such as wind speed, wind direction, air temperature and air pressure of the wind turbine to be detected can be obtained through external sensors such as a wind speed sensor, a wind direction sensor, an air temperature sensor and an air pressure sensor of the wind turbine.

The state parameters and the environment parameters of the wind turbine to be detected can reflect the running condition and the external working condition of the wind turbine to be detected, and a data basis can be provided for fault diagnosis and self-adaptive control of the wind turbine to be detected.

S102, acquiring the fault type and the fault grade of the wind turbine to be detected according to the state parameters and the environment parameters.

In the embodiment of the application, after the state parameters and the environment parameters are acquired, the fault type and the fault grade of the wind turbine to be detected can be acquired according to the state parameters and the environment parameters.

It should be noted that, the specific mode of acquiring the fault type and the fault level of the wind turbine to be detected according to the state parameter and the environment parameter is not limited, and the method can be selected according to actual conditions.

Optionally, the state parameters and the environment parameters can be input into a target fault diagnosis model based on a convolutional neural network (Convolutional Neural Network, abbreviated as CNN) by using a deep learning technology, and the fault type and the fault grade of the wind turbine to be detected are output by the target fault diagnosis model.

Alternatively, the fault type may be a type of fault such as blade cracking, gearbox wear, bearing damage, generator shorting, etc.

Alternatively, the fault level may be set according to actual conditions, for example: mild, moderate, severe, etc.

S103, selecting a target control scheme from the candidate control schemes according to the fault type and the fault level.

In the embodiment of the application, after the fault type and the fault level are obtained, the target control scheme may be selected from the candidate control schemes according to the fault type and the fault level.

Alternatively, the candidate control schemes include a fuzzy logic-based control scheme, a genetic algorithm-based control scheme, and a reinforcement learning-based control scheme.

Alternatively, a mapping relation database between the fault type and the fault level and the control scheme may be pre-established, wherein the mapping relation database includes a control scheme based on fuzzy logic, a control scheme based on genetic algorithm and a control scheme based on reinforcement learning, and the mapping relation database is queried according to the fault type and the fault level to determine the target control scheme.

It should be noted that, by establishing a mapping relation database between the fault type and the fault level and the control scheme, the mapping relation database can be matched with an optimal control scheme according to different fault types and fault levels so as to exert the advantages of each control scheme.

For example, when the failure type of the wind turbine to be detected is a and the failure degree is a, the target control scheme can be determined to be a by querying the mapping relation database.

S104, determining target control parameters and target control strategies of the wind turbine to be detected according to the target control scheme.

The control parameters are parameters affecting the running efficiency and the service life of the wind turbine to be detected.

For example, the control parameters may be mechanical parameters such as blade angle, generator speed, transmission gear ratio, variable blade length, counterweight position, etc.; the generator axial clearance, winding current, exciting current, grid connection parameters and other electrical parameters.

The control strategy is a reasonable control strategy selected according to real-time state parameters of the wind turbine to be detected.

For example, the control strategy may be to select a blade overspeed protection strategy when the wind speed is too high, reduce the energy collection of the wind motor by adjusting the blade angle, select a power limiting output strategy when the power grid demand is low, and control the power generation power by adjusting the electric parameters of the generator terminal.

For example, when the target control scheme is a control scheme based on fuzzy logic, fuzzy reasoning can be performed on the running state and the environment information of the wind turbine to be detected by using the fuzzy logic to obtain target control parameters and a target control strategy; when the target control scheme is a control scheme based on a genetic algorithm, the control parameters and the control strategies of the wind turbine to be detected can be optimized and searched by utilizing the genetic algorithm, so that the target control parameters and the target control strategies are obtained; when the target control scheme is based on reinforcement learning, the control parameters and the control strategies of the wind turbine to be detected can be learned and adjusted online by reinforcement learning, so that the wind turbine to be detected can be ensured to obtain the target control parameters and the target control strategies of the wind turbine to be detected according to different state parameters and environment parameters.

According to the fault diagnosis and control method for the wind turbine, the state parameters and the environment parameters of the wind turbine to be detected are obtained, the fault type and the fault grade of the wind turbine to be detected are obtained according to the state parameters and the environment parameters, the target control scheme is selected from the candidate control schemes according to the fault type and the fault grade, and the target control parameters and the target control strategy of the wind turbine to be detected are determined according to the target control scheme.

In the above embodiment, regarding a specific process of obtaining the fault type and the fault level of the wind turbine to be detected according to the state parameter and the environmental parameter, it can be further understood with reference to fig. 2, and fig. 2 is a schematic flow chart of a fault diagnosis and control method of the wind turbine according to another embodiment of the present application, as shown in fig. 2, the method includes:

S201, obtaining a target fault diagnosis model after training is completed.

It should be noted that, the target fault diagnosis model based on the convolutional neural network CNN provided by the application is constructed on the basis of the traditional convolutional neural network LeNet-5, a convolutional layer, a pooling layer and a full connection layer are added, and a modified linear unit (Rectified linear unit, for short, reLU) function is adopted as an activation function.

The training process of the target failure diagnosis model is explained below.

In the embodiment of the application, normal signals and fault signals of the wind turbine generator under different working conditions can be obtained, the normal signals and the fault signals are randomly divided into a training set and a testing set, the signals in the training set are input into a fault diagnosis model to be trained, the fault diagnosis model is subjected to iterative training by using a cross entropy loss function and a random gradient descent algorithm, a trained fault diagnosis model is obtained, the signals in the testing set are input into the trained fault diagnosis model, the trained fault diagnosis model is responded to meet the training ending condition, and the trained fault diagnosis model is determined to be a target fault diagnosis model.

Optionally, the performance of the target fault diagnosis model may be evaluated using indexes such as accuracy and confusion matrix.

S202, inputting the state parameters and the environment parameters into a target fault diagnosis model, and outputting the fault type and the fault grade of the wind turbine to be detected.

The target fault diagnosis model has the following structure, the first layer of the target fault diagnosis model is an input layer, receives a one-dimensional signal (state parameter and environment parameter) with the length of 2048 as input, the second layer is a convolution layer, carries out convolution operation by using 16 one-dimensional convolution check input signals with the size of 64, the step length of 16 and the filling of 24 to obtain 16 feature images with the length of 128, the third layer is a batch normalization layer (batch normalization layer), carries out normalization processing on the feature images output by the second layer, improves the stability and convergence speed of the model, the fourth layer is an activation layer (activation layer), carries out nonlinear transformation on the feature images output by the third layer by using a ReLU function to enhance the expression capability of the model, the fifth layer is a pooling layer (pooling layer), carries out downsampling processing on the feature images output by the fourth layer by using maximum pooling (max pooling) with the size of 2 and the step length of 2, obtaining 16 feature graphs with the length of 64, the sixth layer to the tenth layer are respectively a convolution layer, a batch normalization layer, an activation layer, a pooling layer and a convolution layer, the sixth layer uses 32 feature graphs with the size of 3, the step length of 1 and the output of a fifth layer of a one-dimensional convolution check with the filling of 1 to carry out convolution operation to obtain 32 feature graphs with the length of 64, carries out batch normalization, reLU activation and maximum pooling operation to the feature graphs to obtain 32 feature graphs with the length of 32, finally uses 64 one-dimensional convolution check with the size of 3, the step length of 1 and the filling of 1 to carry out convolution operation to obtain 64 feature graphs with the length of 32, and the eleventh layer to the fifteenth layer are respectively a batch normalization layer, an activation layer, a pooling layer, a convolution layer and a batch normalization layer, the eleventh layer performs batch normalization, reLU activation and maximum pooling operation on the feature graphs output by the tenth layer to obtain 64 feature graphs with the length of 16, then performs convolution operation on the feature graphs by using 64 one-dimensional convolution cores with the size of 3 and the step length of 1 to obtain 64 feature graphs with the length of 14, finally performs batch normalization operation on the feature graphs, the sixteenth layer to the eighteenth layer are respectively an activation layer, a pooling layer and a full connection layer (fully connected layer), the sixteenth layer performs ReLU activation and maximum pooling operation on the feature graphs output by the fifteenth layer to obtain 64 feature graphs with the length of 7, flattens the feature graphs into a vector with the length of 192, inputs the vector into a full connection layer with 100 neurons to obtain a vector with the length of 100, and the nineteenth layer is a full connection layer, inputs the vector output by the eighteenth layer into a full connection layer with 10 neurons to obtain a vector with the length of 10, namely the fault type and the fault grade of the wind turbine generator.

In order to improve accuracy of determining the fault type and the fault grade of the wind turbine to be detected, cluster analysis is performed on the wind turbine set to determine the final fault type and the final fault grade of the wind turbine to be detected.

In the above embodiment, regarding a specific process of determining the final fault type and fault level of the wind turbine to be detected, it may be further understood with reference to fig. 3, and fig. 3 is a schematic flow chart of a fault diagnosis and control method of a wind turbine according to another embodiment of the present application, as shown in fig. 3, where the method includes:

s301, carrying out cluster analysis on the wind turbine generator set, and dividing the wind turbine generator set into a plurality of subsets according to the state parameter and the environment parameter of each wind turbine generator set.

The wind turbine generator set comprises a plurality of wind turbine generators.

Alternatively, the clustering algorithm (clustering algorithm) may be used, for example: and a K-means algorithm (K-means algorism) is used for carrying out cluster analysis on a plurality of wind turbines, and a wind turbine set is divided into a plurality of subsets according to the state parameter and the environment parameter of each wind turbine.

For example, for wind speed, the wind turbine generator set may be divided into a low wind speed subset, a medium wind speed subset and a high wind speed subset according to the wind speed and the variation range; for temperature, the wind turbine generator set can be divided into a low-temperature subset and a high-temperature subset; for air pressure, the wind turbine generator set can be divided into a high atomic set and a plain subset.

It should be noted that, according to the state parameter and the environmental parameter of each wind turbine, the wind turbine set is divided into a plurality of subsets, so that the states and the environments of the wind turbines in the same subset are similar, and a foundation is laid for subsequent training of a corresponding fault diagnosis model.

S302, training a corresponding target fault diagnosis model aiming at the wind turbine generator in each subset.

After the wind turbine generator set is divided into a plurality of subsets, a corresponding target fault diagnosis model can be trained for the wind turbine generator set in each subset.

It should be noted that, for the specific process of training the target fault diagnosis model, reference may be made to the above embodiment, and details are not repeated here.

For example, for the low temperature subset, a corresponding target fault diagnosis model is trained as M1, and for the high Wen Ziji, a corresponding target fault diagnosis model is trained as M2; for the high atomic set, training a corresponding target fault diagnosis model to be M3, and for the plain subset, training a corresponding target fault diagnosis model to be M4.

S303, inputting the state parameters and the environment parameters of the wind turbine to be detected into all the trained target fault diagnosis models to obtain the fault type and the fault grade output by each target fault diagnosis model.

After training the corresponding target fault diagnosis model for each wind turbine generator in each subset, all trained target fault diagnosis models can be obtained, and the state parameters and the environment parameters of the wind turbine generator to be detected can be input into all trained target fault diagnosis models to obtain the fault type and the fault grade output by each target fault diagnosis model.

S304, processing the fault type and the fault grade output by each target fault diagnosis model through an integrated learning method to obtain the final fault type and the final fault grade of the wind turbine to be detected.

Alternatively, the learning may be integrated by, for example: and comprehensively judging the fault type and the fault grade output by each target fault diagnosis model by a voting method to obtain the final fault type and the final fault grade of the wind turbine to be detected.

According to the fault diagnosis and control method for the wind turbine generator, the target fault diagnosis model is trained through the deep learning technology, the state parameters and the environment parameters are input into the target fault diagnosis model, the fault type and the fault grade of the wind turbine generator to be detected are output, meanwhile, the fault type and the fault grade output by all the target fault diagnosis models are comprehensively judged through the integrated learning method, the final fault type and the final fault grade of the wind turbine generator to be detected are obtained, the accuracy and the high efficiency of fault diagnosis on the wind turbine generator are improved, and a foundation is laid for the follow-up determination of the target control parameters and the target control strategies of the wind turbine generator to be detected.

In the embodiment of the application, after the fault type and the fault grade of the wind turbine to be detected are obtained, a target control scheme can be selected in a self-adaptive mode according to the fault type and the fault grade, and the target control parameters and the target control strategy of the wind turbine to be detected are based on the target control scheme.

The specific process of determining the target control parameters and the target control strategies of the wind turbine to be detected according to the target control scheme provided by the application is explained below.

In the above embodiment, if the target control scheme is a control scheme based on fuzzy logic, with respect to a specific process of determining the target control parameter and the target control policy of the wind turbine to be detected according to the target control scheme, it may be further understood with reference to fig. 4, and fig. 4 is a schematic flow chart of a fault diagnosis and control method of the wind turbine according to another embodiment of the present application, as shown in fig. 4, where the method includes:

s401, determining input variables and output variables of the wind turbine in the fuzzy reasoning process, and setting corresponding fuzzy sets and membership functions for each variable.

The input variable may be wind speed, rotation speed, power, etc., and the output variable may be blade angle.

It should be noted that, for each variable, a corresponding fuzzy set (fuzzy set) and membership function (membership function) may be set, for example: high, medium, low, etc.

For example, for wind speeds, 3 fuzzy sets may be set, 0-3 meters/second as the low wind speed set, 2-5 meters/second as the medium wind speed set, and 4-8 meters/second as the high wind speed set.

It should be noted that "high", "medium" and "low" in the membership functions represent membership of fuzzy sets, i.e. the degree to which a specific numerical value belongs to different fuzzy sets.

For example, for wind speeds, when the wind speed is 2.5 meters/second, the membership to the "low wind speed set" may be 0.2, the membership to the "medium wind speed set" may be 0.8, and the membership to the "high wind speed" may be 0.

S402, inputting real-time input variable values of the wind turbine to be detected into a fuzzy rule base, and performing fuzzy reasoning to obtain corresponding output variable values.

Note that, a fuzzy rule of the wind turbine generator may be preset, for example: "if the wind speed is high and the rotation speed is low, the blade angle is increased", and a preset fuzzy rule is stored in a fuzzy rule library.

In the embodiment of the application, when the real-time input variable value of the wind turbine to be detected is input into the fuzzy rule base, fuzzy reasoning can be performed to obtain the corresponding output variable value.

For example, when the current wind speed is 3 m/s and the rotation speed is 10rpm, according to the preset fuzzy rule of the wind turbine generator, if the wind speed is low and the rotation speed is low, the increment of the blade angle can be obtained to be 2 degrees through fuzzy reasoning by slightly increasing the blade angle.

S403, performing defuzzification on the output variable value, and determining target control parameters and target control strategies of the wind turbine to be detected.

Optionally, fuzzy reasoning results can be summarized to obtain a membership function of the output variable value, and the gravity center of the membership function is calculated and used as a final output value to obtain the target control parameters and the target control strategies of the wind turbine to be detected.

In the above embodiment, if the target control scheme is a control scheme based on a genetic algorithm, with respect to a specific process of determining a target control parameter and a target control policy of a wind turbine to be detected according to the target control scheme, it may be further understood with reference to fig. 5, and fig. 5 is a schematic flow chart of a fault diagnosis and control method of a wind turbine according to another embodiment of the present application, as shown in fig. 5, where the method includes:

s501, coding control parameters and control strategies of the wind turbine generator to obtain a binary character string, wherein the binary character string is used as an individual.

It should be noted that, the control parameters and the control policies of the wind turbine may be encoded into a binary string as an individual (differential).

S502, generating N individuals in a random mode to construct an initial population, and calculating the fitness value of each individual in the initial population, wherein N is a positive integer.

And S503, screening each individual according to the fitness value, forming a new population by a crossing and mutation mode until reaching a preset termination condition, and determining target control parameters and target control strategies of the wind turbine to be detected.

After the fitness value is obtained, the fitness value can be selected, excellent individuals are reserved, poor individuals are eliminated, a new population is formed in a crossing and mutation mode, the steps are repeated until a preset termination condition is reached, and the optimal individuals, namely the target control parameters and the target control strategy, are output.

It should be noted that the setting of the preset termination condition is not limited in the present application, and may be set according to actual situations. Alternatively, the termination condition may be a maximum number of iterations, an optimal solution accuracy, or the like.

In the above embodiment, if the target control scheme is a control scheme based on reinforcement learning, with respect to a specific process of determining the target control parameters and the target control strategies of the wind turbine to be detected according to the target control scheme, it can be further understood with reference to fig. 6, and fig. 6 is a schematic flow chart of a fault diagnosis and control method of the wind turbine according to another embodiment of the present application, as shown in fig. 6, the method includes:

S601, presetting state parameters of a state space of the wind turbine generator, action parameters of an action space and a reward function.

It should be noted that a state space (state space) and an action space (action space) of the wind turbine generator may be set, where state parameters in the state space may include wind speed, rotation speed, power, and the like, and action parameters in the action space may include blade angle, braking force, and the like.

S602, obtaining the rewarding value corresponding to each state parameter-action parameter through the rewarding function.

The method is characterized in that a reward value corresponding to each state parameter-action parameter can be given according to indexes such as running efficiency, reliability and the like of the wind turbine generator through a preset reward function of the wind turbine generator.

S603, selecting target action parameters according to the current state parameters and the rewarding value through a preset strategy function of the wind turbine generator.

It should be noted that, by presetting a policy function of the wind turbine, an optimal or random action can be selected according to the current state parameter and the reward value, so as to explore or utilize the environmental information.

S604, according to the current rewarding value and the future rewarding value, the value corresponding to each state parameter-action parameter is obtained through a preset cost function of the wind turbine generator.

S605, performing iterative updating on the strategy function and the cost function through a reinforcement learning algorithm to obtain a target strategy function and a target cost function.

For example, the objective strategy function and the objective cost function may be obtained by online learning and updating the strategy function and the cost function through reinforcement learning algorithms such as Q learning (Q-learning), SARSA, etc.

S606, determining target control parameters and target control strategies of the wind turbine to be detected according to the target strategy function and the target cost function.

It should be noted that, after the target policy function and the target cost function are obtained, the wind turbine generator may make an optimal control decision according to different state parameters and environment parameters, so as to determine the target control parameter and the target control policy.

According to the fault diagnosis and control method for the wind turbine, after the fault type and the fault level of the wind turbine to be detected are determined, the target control scheme matched with the fault type and the fault level can be determined, and the control parameters and the control strategy of the wind turbine to be detected are dynamically adjusted to obtain the target control parameters and the target control strategy of the wind turbine to be detected, so that the wind turbine is controlled through the target control parameters and the target control strategy, the running efficiency and the reliability of the wind turbine are improved, the abrasion and the loss of the wind turbine are reduced, and the service life of the wind turbine is prolonged.

In the embodiment of the application, after the fault diagnosis data (fault type and fault grade), the target control scheme and the target control strategy of the wind turbine generator are obtained, the data can be stored and shared.

In the above embodiment, regarding the specific process after determining the target control parameters and the target control strategies of the wind turbine to be detected, it may be further understood with reference to fig. 7, and fig. 7 is a schematic flow chart of a fault diagnosis and control method of a wind turbine according to another embodiment of the present application, as shown in fig. 7, where the method includes:

and S701, encrypting, hashing and signing the fault detection data and the experience data of the wind turbine to be detected to form a data block.

The fault diagnosis data and the empirical data include both quantitative operation parameters, signal data, qualitative empirical knowledge, fault diagnosis models, maintenance schemes, and the like.

For example, the fault detection data may include operation data (wind speed, power generation, rotation speed, temperature, etc.), fault data (alarm information, fault code, maintenance record, etc.), wind farm environment data (air temperature, air pressure, humidity, etc.), experience data may include fault mode, fault cause analysis, maintenance advice, etc., fault diagnosis model (model parameters of fault diagnosis model based on deep learning training), maintenance scheme, and maintenance result (process flow and maintenance effect evaluation for various faults).

It should be noted that, by encrypting, hashing and signing (sign) the fault detection data and the experience data of the wind turbine generator to be detected, a data block (block) is formed.

S702, the data block is sent to the block chain network, and verification is carried out on the data block.

Optionally, the data blocks are sent into a blockchain network, which can be verified and agreed upon by multiple nodes to determine the validity and order of the data blocks.

The blockchain technology is a distributed account book technology (distributed ledger technology) which realizes encryption, verification, storage and sharing of data by utilizing a cryptography principle, and has the following characteristics: decentralizing (decentralizing), without relying on any centralized mechanism or platform, data is commonly maintained and updated by multiple nodes; non-tamper (immutability) data cannot be modified or deleted once written into the blockchain, so that the integrity and the authenticity of the data are ensured; traceability (traceability), wherein data is recorded and linked in a block chain according to time sequence to form an indivisible data chain, and the source and the change of the data can be traced; smart contract (smart contract), which refers to an automatic execution protocol based on blockchain technology, can complete transactions or tasks according to preset conditions and rules without manual intervention.

S703, storing the verified data blocks into a blockchain, trading fault detection data and experience data through intelligent contracts on the blockchain, and remotely monitoring and maintaining the wind turbine generator.

The verification-passing data blocks are stored in the block chain, a tamper-proof data chain can be formed, distributed storage and sharing of data are achieved, value evaluation, transaction and excitation are conducted on wind turbine generator fault diagnosis data and experience data through intelligent contracts, value circulation and co-creation of knowledge are achieved, remote monitoring and maintenance are conducted on the wind turbine generator through the intelligent contracts, and functions such as fault early warning, fault processing and fault feedback are achieved.

The application uses the blockchain technology, can encrypt, store and share the failure diagnosis data and the experience data of the wind turbine as a resource in a blockchain network in a distributed manner, realize the decentralized management and circulation of the resource, and simultaneously can use intelligent contracts to realize the excitation mechanism of the failure diagnosis knowledge of the wind turbine, so that related personnel of the wind turbine can create, contribute, verify, evaluate, use and improve the failure diagnosis resource of the wind turbine together to form an open, collaborative and innovative knowledge ecosystem, and can realize the remote monitoring and maintenance of the wind turbine, and automatically trigger corresponding control instructions or maintenance tasks through the intelligent contracts to improve the running efficiency and reliability of the wind turbine.

According to the fault diagnosis and control method for the wind turbine, after the fault diagnosis data, the target control scheme and the target control strategy of the wind turbine are obtained, the block chain technology can be utilized to realize circulation and co-creation of fault diagnosis resources of the wind turbine, remote monitoring and maintenance of the wind turbine are realized, quality and quantity of the fault diagnosis resources of the wind turbine are improved, and running efficiency and reliability of the wind turbine are improved.

The specific process of the fault diagnosis and control method of the wind turbine generator set provided by the application is explained below.

The application takes a 1.5MW doubly-fed induction generator (DFIG) wind turbine generator as a wind turbine generator to be detected, and the wind turbine generator to be detected runs in a mountain wind field for explanation.

It should be noted that, the wind turbine to be detected is internally provided with a temperature sensor, a pressure sensor, a power sensor, a rotation speed sensor and the like, and is externally provided with a wind speed sensor, a wind direction sensor, an air temperature sensor, an air pressure sensor and the like, so that the state parameters and the environment parameters of the wind turbine to be detected can be acquired in real time, and the state parameters and the environment parameters are sent to the cloud server through the wireless communication module.

It should be noted that, when the wind turbine to be detected has a gear box abrasion fault, which causes temperature and pressure increase and power and rotation speed fluctuation in the gear box, the operation efficiency and reliability of the wind turbine to be detected are affected, and the fault is reflected by the acquired state parameters and environmental parameters and is sent to the cloud server.

The method includes the steps that acquired state parameters and environment parameters are input into a target fault diagnosis model to obtain fault types and fault grades of the wind turbine to be detected, the target fault diagnosis model can automatically learn fault characteristics of the wind turbine after a large amount of training of normal and fault signal data, characteristic extraction and classification rules are not required to be set manually, the target fault diagnosis model recognizes that the wind turbine to be detected has a gear box abrasion fault, and the corresponding fault grade is medium.

Further, according to the fault type and the fault level, the selected self-adaptive control method is a target control scheme based on a genetic algorithm, and the control parameters and the control strategies of the wind turbine generator set are optimized and searched by the genetic algorithm, so that the optimal control parameters and the optimal control strategies can be obtained. The method comprises the following specific steps:

Encoding control parameters and control strategies of the wind turbine into a binary character string, wherein the control parameters can be blade angle and braking force, and the control strategies can be braking starting and braking stopping, for example: an individual may be denoted as "00101101", where "00" indicates that the blade angle is 0 ° "10" indicates that the braking force is 50%, "11" indicates that the braking is activated, "01" indicates that the braking is stopped, and a certain number of individuals are randomly generated to form an initial population, alternatively, the population size may be set to 100, that is, 100 individuals are generated, and the fitness value of each individual is calculated.

Alternatively, the fitness value of an individual may be calculated using the following formula:

f＝α×P-β×T-γ×B

wherein f is an fitness value, P is the output power of the wind turbine generator, T is the temperature of the gearbox, B is the braking force, α, β and γ are weight coefficients for adjusting the weight of each performance index, and α=0.8, β=0.1 and γ=0.1 can be taken.

Further, a roulette manner may be adopted to select, that is, the probability of each individual being selected is proportional to the fitness value thereof, select and retain excellent individuals, reject bad individuals, in this example, select by a method of (roulette wire) and mutate according to the crossover probability to generate new individuals, and form a new population, wherein the crossover probability may be set to be 0.8, the mutation probability is 0.01, crossover refers to two individuals exchanging part genes, two new individuals are generated, mutating refers to one individual randomly changing a certain gene, and one new individual is generated, for example, for two individuals "00101101" and "11010010", when crossover is performed, two new individuals "00100010" and "11011101" may be obtained, and when crossover is performed on "00100010", one new individual "00100011" may be obtained.

Repeating the above steps until reaching a preset termination condition, such as a maximum iteration number, an optimal solution precision, etc., where optionally, the preset termination condition may be that the maximum iteration number is 100, that is, repeating the above steps 100 times, and outputting an optimal individual, that is, a target control parameter and a target control policy, for example: the optimal individual is '01011011', namely, the target control parameter is that the blade angle is 15 degrees, the braking force is 75 percent, and the target control strategy is to start braking.

Furthermore, the fault diagnosis data and the experience data of the wind turbine can be encrypted, stored in a distributed mode and shared, circulation and co-creation of fault diagnosis resources of the wind turbine are achieved through a block chain technology, and remote monitoring and maintenance of the wind turbine are achieved. The method comprises the following specific steps:

optionally, an ethernet (ethernet) platform is used as a basic framework of a blockchain network, fault diagnosis data and experience data of the wind turbine are used as a Non-homogeneous Token (NFT) to be issued and traded, wherein the NFT is a digital asset based on a blockchain technology, has the characteristics of uniqueness, irreplaceability, verifiability and the like, can be used for representing any unique thing, and the wind turbine fault diagnosis data and experience data are used as the NFT to be issued and traded, so that the intellectual property, the intellectual traceability, the intellectual anti-counterfeiting and the like of the wind turbine can be ensured.

Through intelligent contracts, an incentive mechanism of wind turbine generator fault diagnosis knowledge is realized, related personnel of the wind turbine generator are encouraged to create, contribute, verify, evaluate, use and improve the wind turbine generator fault diagnosis knowledge together, and an open, collaborative and innovative knowledge ecological system is formed, wherein the incentive mechanism can be set by utilizing the intelligent contracts on the Ethernet platform, such as: aiming at wind turbine generator operation and maintenance personnel, if fault diagnosis data and experience data of the wind turbine generator are reported in time, and verification of other participants is passed, corresponding token rewards can be obtained; aiming at wind turbine generator specialists and researchers, if an effective wind turbine generator fault diagnosis scheme and advice are provided, and corresponding token rewards can be obtained through evaluation of other participants; for other participants, if participating in verification and evaluation of wind turbine fault diagnosis knowledge and providing valuable feedback and improvement comments, corresponding token rewards can be obtained, wherein tokens can be used to purchase or sell other wind turbine fault diagnosis knowledge in a blockchain network or be exchanged for legal money.

Through the excitation mechanism, the value circulation and co-creation of the wind turbine generator fault diagnosis knowledge can be promoted, the quality and quantity of the wind turbine generator fault diagnosis knowledge are improved, an open, collaborative and innovative knowledge ecological system is formed, remote monitoring and maintenance of the wind turbine generator are realized by using a blockchain technology, corresponding control instructions or maintenance tasks are automatically triggered through intelligent contracts, and the running efficiency and reliability of the wind turbine generator are improved.

Optionally, an intelligent contract on the ethernet platform may be used to set a monitoring and maintenance mechanism, for example, for a wind turbine operation and maintenance person, if a control instruction or a maintenance task from a cloud server or a blockchain network is received, the operation may be confirmed and performed according to the instruction or the task through the intelligent contract, after the operation is completed, a result may be reported through the intelligent contract, and a corresponding token reward may be obtained, for a monitoring system in the cloud server or the blockchain network, if a wind turbine is detected to fail or a control parameter and a control policy need to be adjusted, the control instruction or the maintenance task may be sent to the wind turbine operation and maintenance person through the intelligent contract, for the wind turbine itself, if a control instruction or a maintenance task from the cloud server or the blockchain network is received, the operation and maintenance task is automatically executed through the intelligent contract, and the result is reported, if the wind turbine is detected to fail or a control parameter and a control policy need to be adjusted, the wind turbine may be automatically sent to a monitoring system in the cloud server or the blockchain network through the intelligent contract, and adjusted according to feedback, remote monitoring and maintenance of the wind turbine may be realized, and operation efficiency and reliability of the wind turbine may be improved.

In summary, according to the fault diagnosis and control method for the wind turbine generator provided by the application, through a deep learning technology, the trained target fault diagnosis model can automatically learn the fault characteristics of the wind turbine generator, the artificial setting of characteristic extraction and classification rules is not required, the accuracy and the high efficiency of fault diagnosis on the wind turbine generator are improved, the control parameters and the control strategy of the wind turbine generator can be dynamically adjusted according to different fault types and fault grades, the operation efficiency and the reliability of the wind turbine generator are optimized, the abrasion and the loss of the wind turbine generator are reduced, the service life of the wind turbine generator is prolonged, the circulation and the co-creation of fault diagnosis resources of the wind turbine generator and the remote monitoring and the maintenance of the wind turbine generator are realized based on a block chain technology, the quality and the quantity of fault diagnosis resources of the wind turbine generator are improved, and the operation efficiency and the reliability of the wind turbine generator are improved.

Fig. 8 is a schematic structural diagram of a fault diagnosis and control apparatus for a wind turbine according to an embodiment of the present application, as shown in fig. 8, the fault diagnosis and control apparatus 800 for a wind turbine includes a first obtaining module 81, a second obtaining module 82, a selecting module 83, and a determining module 84, where:

the first obtaining module 81 is configured to obtain a state parameter and an environmental parameter of a wind turbine to be detected;

A second obtaining module 82, configured to obtain a fault type and a fault level of the wind turbine to be detected according to the state parameter and the environmental parameter;

a selecting module 83, configured to select a target control scheme from the candidate control schemes according to the fault type and the fault level;

the determining module 84 is configured to determine a target control parameter and a target control policy of the wind turbine to be detected according to the target control scheme.

The fault diagnosis and control device for a wind turbine provided in the second aspect of the present application further has the following technical characteristics, including:

according to one embodiment of the application, the second acquisition module 82 is configured to: obtaining a target fault diagnosis model after training; and inputting the state parameters and the environment parameters into a target fault diagnosis model, and outputting the fault type and the fault grade of the wind turbine to be detected.

According to one embodiment of the application, the second acquisition module 82 is configured to: acquiring normal signals and fault signals of the wind turbine generator under different working conditions, and randomly dividing the normal signals and the fault signals into a training set and a testing set; inputting signals in the training set into a fault diagnosis model to be trained, and performing iterative training on the fault diagnosis model by using a cross entropy loss function and a random gradient descent algorithm to obtain a trained fault diagnosis model; inputting signals in the test set into the trained fault diagnosis model, and determining the trained fault diagnosis model as the target fault diagnosis model in response to the trained fault diagnosis model meeting training ending conditions.

According to one embodiment of the application, the second acquisition module 82 is configured to: performing cluster analysis on a wind turbine generator set, and dividing the wind turbine generator set into a plurality of subsets according to the state parameter and the environment parameter of each wind turbine generator set; training a corresponding target fault diagnosis model aiming at the wind turbine generator in each subset; inputting the state parameters and the environment parameters of the wind turbine to be detected into all the trained target fault diagnosis models to obtain the fault type and the fault grade output by each target fault diagnosis model; and processing the fault type and the fault grade output by each target fault diagnosis model through an integrated learning method to obtain the final fault type and the final fault grade of the wind turbine to be detected.

According to one embodiment of the application, the selection module 83 is configured to: a mapping relation database between the fault type and the fault level and a control scheme is established in advance, wherein the mapping relation database comprises a control scheme based on fuzzy logic, a control scheme based on genetic algorithm and a control scheme based on reinforcement learning; and according to the fault type and the fault level, inquiring the mapping relation database, and determining the target control scheme.

According to one embodiment of the application, the determining module 84 is configured to: determining input variables and output variables of the wind turbine in the fuzzy reasoning process, and setting a corresponding fuzzy set and membership function for each variable; inputting the real-time input variable values of the wind turbine to be detected into a fuzzy rule base, and performing fuzzy reasoning to obtain corresponding output variable values; and performing anti-blurring on the output variable value, and determining target control parameters and target control strategies of the wind turbine to be detected.

According to one embodiment of the application, the determining module 84 is configured to: encoding control parameters and control strategies of the wind turbine generator to obtain a binary character string, wherein the binary character string is used as an individual; generating N individuals in a random mode to construct an initial population, and calculating the fitness value of each individual in the initial population, wherein N is a positive integer; and screening each individual according to the fitness value, forming a new population in a crossing and mutation mode until a preset termination condition is reached, and determining target control parameters and target control strategies of the wind turbine to be detected.

According to one embodiment of the application, the determining module 84 is configured to: presetting state parameters of a state space of a wind turbine generator, action parameters of an action space and a reward function; obtaining a reward value corresponding to each state parameter-action parameter through the reward function; selecting target action parameters according to the current state parameters and the rewarding values through a preset strategy function of the wind turbine generator; according to the current rewarding value and the future rewarding value, the value corresponding to each state parameter-action parameter is obtained through a preset cost function of the wind turbine generator; performing iterative updating on the strategy function and the cost function through a reinforcement learning algorithm to obtain a target strategy function and a target cost function; and determining target control parameters and target control strategies of the wind turbine to be detected according to the target strategy function and the target cost function.

According to one embodiment of the application, the apparatus 800 is for: encrypting, hashing and signing fault detection data and experience data of the wind turbine to be detected to form a data block; transmitting the data block to a blockchain network, and verifying the data block; and storing the verified data blocks into a blockchain, trading the fault detection data and the experience data through intelligent contracts on the blockchain, and remotely monitoring and maintaining the wind turbine generator.

According to the fault diagnosis and control device for the wind turbine, the state parameters and the environment parameters of the wind turbine to be detected are obtained, the fault type and the fault grade of the wind turbine to be detected are obtained according to the state parameters and the environment parameters, the target control scheme is selected from the candidate control schemes according to the fault type and the fault grade, and the target control parameters and the target control strategy of the wind turbine to be detected are determined according to the target control scheme.

To achieve the above embodiments, the present application also provides an electronic device, a computer-readable storage medium, and a computer program product.

Fig. 9 is a block diagram of an electronic device according to an embodiment of the present application, and as shown in fig. 9, a device 1000 includes a memory 101, a processor 102, and a computer program stored on the memory 101 and executable on the processor 102, where the processor 102 executes program instructions to implement a fault diagnosis and control method for executing the wind turbine generator according to the embodiment of fig. 1 to 7.

In order to implement the above-described embodiments, the present application also provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the fault diagnosis and control method of the wind turbine generator set of the embodiments of fig. 1 to 7.

In order to implement the above embodiments, the present application also provides a computer program product which, when executed by an instruction processor in the computer program product, performs the method of fault diagnosis and control of a wind turbine generator of the embodiments of fig. 1 to 7.

In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.

Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present application, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise.

Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and additional implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order from that shown or discussed, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the embodiments of the present application.

Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.

It is to be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. As with the other embodiments, if implemented in hardware, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.

Those of ordinary skill in the art will appreciate that all or a portion of the steps carried out in the method of the above-described embodiments may be implemented by a program to instruct related hardware, where the program may be stored in a computer readable storage medium, and where the program, when executed, includes one or a combination of the steps of the method embodiments.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing module, or each unit may exist alone physically, or two or more units may be integrated in one module. The integrated modules may be implemented in hardware or in software functional modules. The integrated modules may also be stored in a computer readable storage medium if implemented in the form of software functional modules and sold or used as a stand-alone product.

The above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, or the like. While embodiments of the present application have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the application, and that variations, modifications, alternatives and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the application.

Claims (10)

1. A method for fault diagnosis and control of a wind turbine, the method comprising:

acquiring state parameters and environment parameters of a wind turbine to be detected;

acquiring the fault type and the fault grade of the wind turbine to be detected according to the state parameters and the environment parameters;

selecting a target control scheme from the candidate control schemes according to the fault type and the fault level;

and determining target control parameters and target control strategies of the wind turbine to be detected according to the target control scheme.

2. The method according to claim 1, wherein the obtaining the fault type and the fault level of the wind turbine to be detected according to the state parameter and the environmental parameter includes:

obtaining a target fault diagnosis model after training;

And inputting the state parameters and the environment parameters into a target fault diagnosis model, and outputting the fault type and the fault grade of the wind turbine to be detected.

3. The method of claim 2, wherein the obtaining a trained target fault diagnosis model comprises:

acquiring normal signals and fault signals of the wind turbine generator under different working conditions, and randomly dividing the normal signals and the fault signals into a training set and a testing set;

inputting signals in the training set into a fault diagnosis model to be trained, and performing iterative training on the fault diagnosis model by using a cross entropy loss function and a random gradient descent algorithm to obtain a trained fault diagnosis model;

inputting signals in the test set into the trained fault diagnosis model, and determining the trained fault diagnosis model as the target fault diagnosis model in response to the trained fault diagnosis model meeting training ending conditions.

4. A method according to any one of claims 2-3, wherein said obtaining the fault type and the fault level of the wind turbine to be detected based on the state parameter and the environmental parameter comprises:

Performing cluster analysis on a wind turbine generator set, and dividing the wind turbine generator set into a plurality of subsets according to the state parameter and the environment parameter of each wind turbine generator set;

training a corresponding target fault diagnosis model aiming at the wind turbine generator in each subset;

inputting the state parameters and the environment parameters of the wind turbine to be detected into all the trained target fault diagnosis models to obtain the fault type and the fault grade output by each target fault diagnosis model;

and processing the fault type and the fault grade output by each target fault diagnosis model through an integrated learning method to obtain the final fault type and the final fault grade of the wind turbine to be detected.

5. The method of claim 1, wherein selecting a target control scheme from candidate control schemes according to the fault type and fault level comprises:

a mapping relation database between the fault type and the fault level and a control scheme is established in advance, wherein the mapping relation database comprises a control scheme based on fuzzy logic, a control scheme based on genetic algorithm and a control scheme based on reinforcement learning;

And according to the fault type and the fault level, inquiring the mapping relation database, and determining the target control scheme.

6. The method according to claim 5, wherein if the target control scheme is a fuzzy logic based control scheme, the determining, according to the target control scheme, the target control parameters and the target control strategy of the wind turbine to be detected includes:

determining input variables and output variables of the wind turbine in the fuzzy reasoning process, and setting a corresponding fuzzy set and membership function for each variable;

inputting the real-time input variable values of the wind turbine to be detected into a fuzzy rule base, and performing fuzzy reasoning to obtain corresponding output variable values;

and performing anti-blurring on the output variable value, and determining target control parameters and target control strategies of the wind turbine to be detected.

7. The method according to claim 5, wherein if the target control scheme is a control scheme based on a genetic algorithm, the determining, according to the target control scheme, a target control parameter and a target control strategy of the wind turbine to be detected includes:

encoding control parameters and control strategies of the wind turbine generator to obtain a binary character string, wherein the binary character string is used as an individual;

Generating N individuals in a random mode to construct an initial population, and calculating the fitness value of each individual in the initial population, wherein N is a positive integer;

and screening each individual according to the fitness value, forming a new population in a crossing and mutation mode until a preset termination condition is reached, and determining target control parameters and target control strategies of the wind turbine to be detected.

8. The method according to claim 5, wherein if the target control scheme is a reinforcement learning-based control scheme, the determining target control parameters and target control strategies of the wind turbine to be detected according to the target control scheme includes:

presetting state parameters of a state space of a wind turbine generator, action parameters of an action space and a reward function;

obtaining a reward value corresponding to each state parameter-action parameter through the reward function;

selecting target action parameters according to the current state parameters and the rewarding values through a preset strategy function of the wind turbine generator;

according to the current rewarding value and the future rewarding value, the value corresponding to each state parameter-action parameter is obtained through a preset cost function of the wind turbine generator;

Performing iterative updating on the strategy function and the cost function through a reinforcement learning algorithm to obtain a target strategy function and a target cost function;

and determining target control parameters and target control strategies of the wind turbine to be detected according to the target strategy function and the target cost function.

9. The method according to claim 1, characterized in that the method further comprises:

encrypting, hashing and signing fault detection data and experience data of the wind turbine to be detected to form a data block;

transmitting the data block to a blockchain network, and verifying the data block;

and storing the verified data blocks into a blockchain, trading the fault detection data and the experience data through intelligent contracts on the blockchain, and remotely monitoring and maintaining the wind turbine generator.

10. A fault diagnosis and control device for a wind turbine, the device comprising:

the first acquisition module is used for acquiring state parameters and environment parameters of the wind turbine to be detected;

the second acquisition module is used for acquiring the fault type and the fault grade of the wind turbine to be detected according to the state parameters and the environment parameters;

The selecting module is used for selecting a target control scheme from the candidate control schemes according to the fault type and the fault level;

and the determining module is used for determining target control parameters and target control strategies of the wind turbine to be detected according to the target control scheme.