Operation maintenance management system and method for wind power generation equipment

Abstract

The invention relates to the technical field of intelligent management, and particularly discloses an operation maintenance management system and method for wind power generation equipment, and the system employs an artificial intelligence technology based on deep learning to monitor and analyze the operation data and environment data of the wind power generation equipment, and captures the operation state time sequence change characteristics of the wind power generation equipment. And whether the performance of the equipment is abnormal or not is intelligently judged based on the time sequence response characteristics of the operation state of the wind power generation equipment relative to the environment data. Therefore, real-time monitoring, performance evaluation and intelligent maintenance of the wind power generation equipment can be realized, so that the operation efficiency and stability of the wind power generation equipment are improved.

Description

Technical Field

[0001] This application relates to the field of intelligent management technology, and more specifically, to an operation and maintenance management system and method for wind power generation equipment. Background Technology

[0002] Wind power generation refers to the process of converting the kinetic energy of wind into electrical energy. Wind power generation equipment utilizes the power of wind to drive a generator through the rotation of the turbine blades of a wind turbine generator set. Specifically, when wind blows onto the turbine blades of a wind turbine generator set, due to the special shape of the blades, some of the wind's kinetic energy is converted into the rotational kinetic energy of the blades. This rotational kinetic energy further drives the generator connected to the turbine to rotate, thus converting the kinetic energy into electrical energy. With the increasing demand for renewable energy in society, wind power generation, as a green and environmentally friendly energy supply method, has been widely used globally.

[0003] Because wind power equipment is typically installed in open outdoor environments, it is inevitably affected by natural factors such as wind, temperature, and humidity. To ensure the stability of the equipment, operation and maintenance management are necessary. However, traditional methods of wind power equipment maintenance usually rely on manual inspections and periodic maintenance. This approach is not only inefficient but also suffers from problems such as untimely and inaccurate monitoring, making real-time monitoring and early warning difficult. Furthermore, since wind power equipment is often installed in remote areas, real-time monitoring and maintenance become particularly challenging.

[0004] Therefore, there is a need for an optimized operation and maintenance management system and method for wind power generation equipment. Summary of the Invention

[0005] To address the aforementioned technical problems, this application is proposed. Embodiments of this application provide an operation and maintenance management system and method for wind power generation equipment. This system utilizes deep learning-based artificial intelligence technology to monitor and analyze the operational and environmental data of the wind power generation equipment, capturing the temporal variation characteristics of the wind power generation equipment's operational status and the temporal variation characteristics of the environmental data. Based on the temporal response characteristics of the wind power generation equipment's operational status relative to the environmental data, it intelligently determines whether there are any performance anomalies in the equipment. This enables real-time monitoring, performance evaluation, and intelligent maintenance of the wind power generation equipment, thereby improving its operational efficiency and stability.

[0006] Accordingly, according to one aspect of this application, an operation and maintenance management system for wind power generation equipment is provided, comprising:

[0007] The wind power generation information acquisition module is used to acquire the operating data of the wind power generation equipment at multiple predetermined time points within a predetermined time period, as well as the environmental data at multiple predetermined time points within the predetermined time period. The operating data includes the power of the equipment, the wind turbine speed, the current and the temperature, and the environmental data includes the wind speed, the wind direction and the air pressure.

[0008] The device operation status timing feature extraction module is used to perform timing association encoding on the operation data at the multiple predetermined time points to obtain the device operation status timing feature vector;

[0009] An environmental temporal feature extraction module is used to extract temporal features from the environmental data at the multiple predetermined time points to obtain an environmental data temporal feature vector.

[0010] The temporal correlation response module is used to perform correlation encoding and correlation feature enhancement on the temporal feature vector of the device operating state and the temporal feature vector of the environmental data to obtain a salient device operating state-environment temporal response feature matrix.

[0011] The equipment performance analysis module is used to determine whether there are any abnormalities in the performance of the wind power generation equipment based on the salienced equipment operating status-environment time-series response feature matrix.

[0012] In the aforementioned wind power generation equipment operation and maintenance management system, the equipment operation status time sequence feature extraction module includes: an operation data normalization unit, used to normalize the operation data at the multiple predetermined time points to obtain an operation data input matrix; and an operation status time sequence feature extraction unit, used to pass the operation data input matrix through a convolutional neural network-based equipment operation status time sequence association encoder to obtain the equipment operation status time sequence feature vector.

[0013] In the above-mentioned wind power generation equipment operation and maintenance management system, the data normalization unit is used to: arrange the operation data of the multiple predetermined time points into the operation data input matrix according to the time dimension and the sample dimension.

[0014] In the above-mentioned wind power generation equipment operation and maintenance management system, the environmental time series feature extraction module includes: an environmental data regularization unit, used to arrange the environmental data of the multiple predetermined time points into an environmental data input matrix according to the time dimension and the sample dimension; and an environmental time series feature extraction unit, used to pass the environmental data input matrix through an environmental time series feature extractor based on a convolutional neural network to obtain the environmental data time series feature vector.

[0015] In the aforementioned wind power generation equipment operation and maintenance management system, the time-series correlation response module includes: a correlation unit, used to correlate and encode the equipment operation state time-series feature vector and the environmental data time-series feature vector to obtain an equipment operation state-environment time-series response feature matrix; and a feature saliency unit, used to pass the equipment operation state-environment time-series response feature matrix through a feature saliency unit based on a spatial attention layer to obtain the saliency equipment operation state-environment time-series response feature matrix.

[0016] In the aforementioned wind power generation equipment operation and maintenance management system, the association unit includes: a dynamic optimization subunit, used to perform differential entropy quantization dynamic optimization on the equipment operating state time-series feature vector and the environmental data time-series feature vector respectively to obtain an optimized equipment operating state time-series feature vector and an optimized environmental data time-series feature vector; a feature fusion subunit, used to perform weighted fusion on the optimized equipment operating state time-series feature vector and the optimized environmental data time-series feature vector to obtain an equipment operating state-environment time-series response feature vector; and an autocorrelation subunit, used to multiply the equipment operating state-environment time-series response feature vector with its own transpose to obtain the equipment operating state-environment time-series response feature matrix.

[0017] In the aforementioned wind power equipment operation and maintenance management system, the dynamic optimization subunit is used to: calculate the self-similarity matrix of the equipment operating state time-series feature vectors, and perform key dimension reduction on the self-similarity matrix of the equipment operating state time-series feature vectors to obtain a set of inherent component encoding vectors of the equipment operating state time-series feature vectors; input the set of inherent component encoding vectors of the equipment operating state time-series feature vectors into a sequence encoder based on a forward LSTM model to obtain a set of inherent component context-related encoding vectors of the equipment operating state time-series feature vectors; and calculate the inherent component encoding vectors of the equipment operating state time-series feature vectors. The set of bitwise fluctuation entropy is obtained by dividing the set of context-associative encoded vectors and the set of encoded vectors of inherent components of the device operating state temporal feature vectors. The set of bitwise fluctuation entropy is then weighted using the Softmax function to obtain a set of bitwise transformation entropy adjustment parameters. Based on the set of bitwise transformation entropy adjustment parameters, the set of device operating state temporal feature vectors is fused to obtain an optimized device operating state temporal feature vector.

[0018] In the aforementioned wind power generation equipment operation and maintenance management system, the feature saliency unit includes: a deep convolutional encoding subunit, used to perform deep convolutional encoding on the equipment operating state-environment time-series response feature matrix using the convolutional layer of the feature saliency unit to obtain an initial convolutional feature map; a spatial attention generation subunit, used to input the initial convolutional feature map into the spatial attention branch of the spatial attention mechanism model to obtain a spatial attention map; an activation subunit, used to apply the spatial attention map to the Softmax activation function to obtain a spatial attention feature map; a spatial attention application subunit, used to calculate the positional dot product of the spatial attention feature map and the initial convolutional feature map to obtain a spatially enhanced equipment operating state-environment time-series response feature map; and a pooling subunit, used to perform mean pooling along the channel dimension on the spatially enhanced equipment operating state-environment time-series response feature map to obtain the saliency equipment operating state-environment time-series response feature matrix.

[0019] In the aforementioned wind power equipment operation and maintenance management system, the equipment performance analysis module is used to: pass the saliency equipment operating status-environment time-series response feature matrix through a classifier to obtain a classification result, and the classification result is used to determine whether the performance of the wind power equipment is abnormal.

[0020] According to another aspect of this application, a method for operation, maintenance and management of wind power generation equipment is provided, comprising:

[0021] The system acquires operational data of a wind power generation device at multiple predetermined time points within a predetermined time period, as well as environmental data at multiple predetermined time points within the predetermined time period. The operational data includes the device's power, rotor speed, current, and temperature, while the environmental data includes wind speed, wind direction, and air pressure.

[0022] The operation data at the multiple predetermined time points are time-series correlated and encoded to obtain the time-series feature vector of the device operation status;

[0023] Temporal feature extraction is performed on the environmental data at the multiple predetermined time points to obtain the environmental data temporal feature vector;

[0024] The temporal feature vectors of the device operating status and the temporal feature vectors of the environmental data are correlated and encoded, and the correlation features are enhanced to obtain a salientized device operating status-environment temporal response feature matrix.

[0025] Based on the saliency of the equipment operating status-environment time-series response feature matrix, it is determined whether there are any abnormalities in the performance of the wind power generation equipment.

[0026] Compared with existing technologies, the wind power generation equipment operation and maintenance management system and method provided in this application utilize deep learning-based artificial intelligence technology to monitor and analyze the operation data and environmental data of the wind power generation equipment. It captures the temporal variation characteristics of the wind power generation equipment's operating status and the environmental data, and intelligently judges whether there are any performance abnormalities based on the temporal response characteristics of the wind power generation equipment's operating status relative to the environmental data. This enables real-time monitoring, performance evaluation, and intelligent maintenance of the wind power generation equipment, thereby improving its operating efficiency and stability. Attached Figure Description

[0027] The above and other objects, features, and advantages of this application will become more apparent from the more detailed description of the embodiments of this application in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain this application and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0028] Figure 1 This is a block diagram of the operation and maintenance management system for wind power generation equipment according to an embodiment of this application.

[0029] Figure 2 This is a schematic diagram of the architecture of the operation and maintenance management system for wind power generation equipment according to an embodiment of this application.

[0030] Figure 3 This is a block diagram of the equipment operation status timing feature extraction module in the wind power generation equipment operation and maintenance management system according to an embodiment of this application.

[0031] Figure 4 This is a block diagram of the environmental time-series feature extraction module in the operation and maintenance management system of wind power generation equipment according to an embodiment of this application.

[0032] Figure 5 This is a block diagram of the timing-related response module in the operation and maintenance management system for wind power generation equipment according to an embodiment of this application.

[0033] Figure 6 This is a flowchart of a method for the operation, maintenance and management of wind power generation equipment according to an embodiment of this application. Detailed Implementation

[0034] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0035] Figure 1This is a block diagram of the operation and maintenance management system for wind power generation equipment according to an embodiment of this application.

[0036] Figure 2 This is a schematic diagram of the architecture of an operation and maintenance management system for wind power generation equipment according to an embodiment of this application. Figure 1 and Figure 2 As shown, the wind power generation equipment operation and maintenance management system 100 according to an embodiment of this application includes: a wind power generation information acquisition module 110, used to acquire the operation data of the wind power generation equipment at multiple predetermined time points within a predetermined time period, and the environmental data at multiple predetermined time points within the predetermined time period, wherein the operation data includes the equipment's power, rotor speed, current, and temperature, and the environmental data includes wind speed, wind direction, and air pressure; an equipment operation status time-series feature extraction module 120, used to perform time-series correlation encoding on the operation data at the multiple predetermined time points to obtain an equipment operation status time-series feature vector; an environmental time-series feature extraction module 130, used to perform time-series feature extraction on the environmental data at the multiple predetermined time points to obtain an environmental data time-series feature vector; a time-series correlation response module 140, used to perform correlation encoding and correlation feature enhancement on the equipment operation status time-series feature vector and the environmental data time-series feature vector to obtain a saliency equipment operation status-environment time-series response feature matrix; and an equipment performance analysis module 150, used to determine whether there is an abnormality in the performance of the wind power generation equipment based on the saliency equipment operation status-environment time-series response feature matrix.

[0037] As mentioned in the background section, wind power is a green and renewable energy conversion method. However, traditional operation and maintenance management methods often rely on manual inspections and periodic maintenance, which are inefficient and lack real-time monitoring and early warning capabilities. Furthermore, wind power equipment is often installed in remote areas, such as mountainous regions and coastal areas, where harsh climate conditions pose significant challenges to real-time monitoring and maintenance.

[0038] To address the aforementioned technical problems, the technical concept of this application is to utilize deep learning-based artificial intelligence technology to monitor and analyze the operational and environmental data of wind power generation equipment. This involves capturing the temporal variation characteristics of the wind power generation equipment's operational status and the environmental data, and intelligently determining whether there are any performance anomalies based on the temporal response characteristics of the wind power generation equipment's operational status relative to the environmental data. This enables real-time monitoring, performance evaluation, and intelligent maintenance of wind power generation equipment, thereby improving its operational efficiency and stability.

[0039] In the aforementioned wind power equipment operation and maintenance management system 100, the wind power information acquisition module 110 is used to acquire operational data of the wind power equipment at multiple predetermined time points within a predetermined time period, as well as environmental data at multiple predetermined time points within the predetermined time period. The operational data includes the equipment's power, rotor speed, current, and temperature, while the environmental data includes wind speed, wind direction, and air pressure. It should be understood that wind power equipment utilizes the power of the wind to drive the rotation of the wind turbine blades, thereby driving a generator to produce electrical energy. In other words, the power generation efficiency of wind power equipment is closely related to meteorological factors. Therefore, in the technical solution of this application, the working state and power generation efficiency of the wind power equipment are characterized by acquiring power, rotor speed, current, and temperature data during its operation. Simultaneously, corresponding wind speed, wind direction, and air pressure data are acquired. By comprehensively analyzing the equipment's operational data and environmental data, the correlation between the equipment's operating state and environmental factors is established, thereby identifying abnormal performance conditions of the equipment.

[0040] In the aforementioned wind power equipment operation and maintenance management system 100, the equipment operation status time-series feature extraction module 120 is used to perform time-series correlation encoding on the operation data at multiple predetermined time points to obtain the equipment operation status time-series feature vector. Specifically, Figure 3 This is a block diagram of the equipment operation status timing feature extraction module in the wind power generation equipment operation and maintenance management system according to an embodiment of this application. Figure 3 As shown, the device operation status time sequence feature extraction module 120 includes: an operation data normalization unit 121, used to normalize the operation data at the multiple predetermined time points to obtain an operation data input matrix; and an operation status time sequence feature extraction unit 122, used to pass the operation data input matrix through a device operation status time sequence association encoder based on a convolutional neural network to obtain the device operation status time sequence feature vector.

[0041] Specifically, the operation data normalization unit 121 is used to normalize the operation data at the multiple predetermined time points to obtain an operation data input matrix. In a specific example of this application, the processing method for normalizing the operation data at the multiple predetermined time points to obtain the operation data input matrix is ​​to arrange the operation data at the multiple predetermined time points according to the time dimension and the sample dimension to form the operation data input matrix. It should be understood that the operation data of wind power generation equipment often has time dependence, that is, the data at the previous moment may affect the state at the subsequent moment. Therefore, in order to retain the temporal sequence information of the operation data of wind power generation equipment and consider the time dependence between the operation data at different time points, the operation data at the multiple predetermined time points are further arranged according to the time dimension and the sample dimension to integrate the temporal correlation information between various operation parameters, thereby facilitating the capture of the dynamic change characteristics of the operation state of wind power generation equipment over time and improving the model's ability to understand the operation state of wind power generation equipment.

[0042] Specifically, the operating state temporal feature extraction unit 122 is used to process the operating data input matrix through a device operating state temporal correlation encoder based on a convolutional neural network to obtain the device operating state temporal feature vector. It should be understood that a convolutional neural network (CNN) is a feedforward neural network that can automatically learn feature information in input data through backpropagation. In the technical solution of this application, a device operating state temporal correlation encoder based on a convolutional neural network is used to process the operating data input matrix. This encoder can utilize the convolutional layers of the convolutional neural network model to extract the local structural patterns of the operating data input matrix using sliding convolution operations, mining the local correlation information of various operating parameters in the time dimension, better capturing the temporal correlation between device operating data, and helping to understand the evolution process of device operating states. Simultaneously, by utilizing the pooling layers and fully connected layers of the convolutional neural network model, the device operating state temporal correlation encoder can achieve feature dimensionality reduction and abstraction, helping to abstract more discriminative feature representations from the operating data input matrix, reducing the impact of data dimensionality and noise, lowering the risk of overfitting, and improving the model's generalization ability.

[0043] In the aforementioned wind power equipment operation and maintenance management system 100, the environmental time-series feature extraction module 130 is used to extract time-series features from environmental data at multiple predetermined time points to obtain environmental data time-series feature vectors. Specifically, Figure 4 This is a block diagram of the environmental time-series feature extraction module in the operation and maintenance management system of wind power generation equipment according to an embodiment of this application. Figure 4As shown, the environmental temporal feature extraction module 130 includes: an environmental data normalization unit 131, used to arrange the environmental data at multiple predetermined time points into an environmental data input matrix according to the time dimension and the sample dimension; and an environmental temporal feature extraction unit 132, used to pass the environmental data input matrix through an environmental temporal feature extractor based on a convolutional neural network to obtain the environmental data temporal feature vector.

[0044] Specifically, the environmental data normalization unit 131 is used to arrange the environmental data at multiple predetermined time points into an environmental data input matrix according to the time dimension and the sample dimension. It should be understood that the environmental data at multiple predetermined time points also have time dependence. Therefore, similarly, arranging the environmental data at multiple predetermined time points according to the time dimension and the sample dimension preserves the temporal order of the environmental data and integrates the temporal correlation information between various environmental parameters, thereby better representing the dynamic changes in the operating environment of wind power generation equipment and improving the model's ability to understand environmental factors.

[0045] Specifically, the environmental temporal feature extraction unit 132 is used to process the environmental data input matrix through an environmental temporal feature extractor based on a convolutional neural network to obtain the environmental data temporal feature vector. It should be understood that a convolutional neural network (CNN) can automatically learn feature information from input data. Similarly, in the technical solution of this application, an environmental temporal feature extractor based on a convolutional neural network is used to process the environmental data input matrix. By performing convolution and pooling operations on the environmental data input matrix, the environmental temporal feature extractor can extract the local feature representation of the environmental data input matrix, uncover the local correlation information of various environmental parameters in the time dimension, and thus better understand the changing process of the operating environment of wind power generation equipment.

[0046] In the aforementioned wind power equipment operation and maintenance management system 100, the time-series correlation response module 140 is used to perform correlation encoding and correlation feature enhancement on the time-series feature vector of the equipment operating status and the time-series feature vector of the environmental data to obtain a salient equipment operating status-environment time-series response feature matrix. Specifically, Figure 5 This is a block diagram of the timing-related response module in the operation and maintenance management system for wind power generation equipment according to an embodiment of this application. Figure 5 As shown, the time-series correlation response module 140 includes: a correlation unit 141, used to correlate and encode the device operating state time-series feature vector and the environmental data time-series feature vector to obtain a device operating state-environment time-series response feature matrix; and a feature saliency unit 142, used to pass the device operating state-environment time-series response feature matrix through a feature saliency generator based on a spatial attention layer to obtain a saliency device operating state-environment time-series response feature matrix.

[0047] Specifically, the association unit 141 is used to perform association encoding on the time-series feature vector of the equipment operating state and the time-series feature vector of the environmental data to obtain the equipment operating state-environment time-series response feature matrix. It should be understood that the operating state of wind power generation equipment is often affected by environmental conditions. That is, there is a certain correlation response relationship between the operating state of wind power generation equipment and environmental data. Therefore, in order to extract the time-series response relationship between the equipment operating state and environmental factors for evaluating the performance of wind power generation equipment, the time-series feature vector of the equipment operating state and the time-series feature vector of the environmental data are further association encoded.

[0048] Specifically, the association unit 141 includes: a dynamic optimization subunit, used to perform differential entropy quantization dynamic optimization on the device operating state time-series feature vector and the environmental data time-series feature vector respectively to obtain an optimized device operating state time-series feature vector and an optimized environmental data time-series feature vector; a feature fusion subunit, used to perform weighted fusion on the optimized device operating state time-series feature vector and the optimized environmental data time-series feature vector to obtain a device operating state-environment time-series response feature vector; and an autocorrelation subunit, used to multiply the device operating state-environment time-series response feature vector with its own transpose to obtain the device operating state-environment time-series response feature matrix.

[0049] More specifically, the dynamic optimization subunit is used to: calculate the self-similarity matrix of the device operating state temporal feature vector, and perform key dimension reduction on the self-similarity matrix of the device operating state temporal feature vector to obtain a set of inherent component encoding vectors of the device operating state temporal feature vector, expressed by the formula:

[0050]

[0051] Where V represents the timing feature vector of the device's operating state, T represents the transpose of the vector, and M... z Let v1, v2, v3 represent the self-similarity matrix, U represent the set of encoding vectors of the inherent components of the temporal features of the device's operating state, and v1, v2, v3 represent the self-similarity matrix. m Let λ1, λ2, and λ3 represent the encoding vectors of the inherent components of the temporal features of the first, second, and m-th devices, respectively. Let Λ represent the diagonal matrix of the temporal features of the devices' operational status after dimensionality reduction of the key dimensions. m These represent the first and m-th eigenvalues ​​on the diagonal of the time-series feature matrix of the device's operating status, respectively.

[0052] In other words, by calculating the self-similarity matrix, the correlation strength between the temporal feature vectors of equipment operating status at different time lags can be quantified, revealing their dynamic coupling patterns, such as the temporal synchronicity between power fluctuations and wind speed changes. This analysis provides a foundation for subsequent feature dimensionality reduction, while the introduction of Principal Process Analysis (PCA) solves the problems of low computational efficiency and noise interference caused by excessive redundant information in high-dimensional data. Specifically, PCA extracts the principal components in the self-similarity matrix, i.e., the orthogonal directions with the largest variance, mapping the temporal feature vectors of equipment operating status to a low-dimensional space while retaining the most core temporal change patterns in the data. This strengthens the significant correlation features between equipment operating status and environmental response, thereby improving the sensitivity and accuracy of anomaly detection.

[0053] More specifically, the dynamic optimization subunit is further configured to: input the set of encoded vectors of the inherent components of the device operating state temporal feature vector into a sequence encoder based on a forward LSTM model to obtain the set of encoded vectors of the context association of the inherent components of the device operating state temporal feature vector, expressed by the formula:

[0054] F = LSTM([v1,v2,…,v...) m ])=[s1,s2,…,s m ]

[0055] Where LSTM represents the forward LSTM model, F represents the set of context-related encoding vectors of the inherent components of the temporal features of the device's operating state, and s1, s2, s... m This represents the context-related encoding vector of the inherent components of the temporal features of the operating status of the first, second, and m-th devices.

[0056] In other words, by introducing a forward LSTM model, the temporal dependencies between the inherent component encoding vectors of the temporal feature vectors of each device's operating state can be captured layer by layer using sequence modeling, especially long-term contextual associations. Specifically, by dynamically integrating historical contextual information through the gating mechanism of LSTM, the set of discrete device operating state temporal feature vector inherent component encoding vectors is transformed into a set of device operating state temporal feature inherent component contextual association encoding vectors with strong temporal semantics. This not only retains the core feature expression capabilities after PCA processing but also enhances the representation accuracy of nonlinear dynamic interactions, significantly improving the comprehensiveness and accuracy of device performance anomaly detection.

[0057] More specifically, the dynamic optimization subunit is further configured to: calculate the bit-by-bit fluctuation entropy between the context-associated encoding vectors of the inherent components of the device operating state temporal feature vectors and the encoding vectors of the inherent components of the device operating state temporal feature vectors for each group in the set of the device operating state temporal feature vectors and the encoding vectors of the inherent components of the device operating state temporal feature vectors, to obtain a set of bit-by-bit fluctuation entropies, expressed by the formula:

[0058]

[0059] Where ∧ represents a logical operator, w represents a bitwise comparison function, and v i This represents the encoding vector of the inherent components of the time-series features of the operating status of the i-th device. s represents the feature value at the j-th position of the encoding vector of the intrinsic components of the time-series features of the i-th device's operating state. i This represents the context-related encoding vector of the intrinsic components of the time-series features of the operating state of the i-th device. Let represent the feature value at the j-th position of the context-associative encoding vector of the intrinsic components of the operating state temporal feature of the i-th device. ε represents a predetermined threshold, which can be set to 1.2. However, this example does not impose a specific limit and can be adjusted according to actual conditions. i This represents the i-th position-by-position fluctuation matching feature vector. Let L represent the eigenvalue at the j-th position of the i-th position-by-position oscillation matching eigenvector, and let e represent the length of the i-th position-by-position oscillation matching eigenvector. i Let represent the bitwise fluctuation entropy of the i-th position.

[0060] In other words, by calculating bit-by-bit fluctuation entropy, the set of context-related encoded vectors of the inherent components of the device operating state temporal feature vector and the bit-level differences of the encoded vectors in binary encoding can be quantified. This captures the specific patterns of information gain or perturbation. High bit entropy values ​​may indicate drastic changes in key information bits, while low entropy values ​​may reflect the stability of redundant information. In this way, feature differences are transformed into quantifiable entropy signals, thus providing a basis for differentiated weight allocation for subsequent dynamic weight adjustment and feature fusion. This significantly improves the granularity of feature difference representation and enhances the sensitivity of anomaly detection and the ability to suppress false alarms.

[0061] More specifically, the dynamic optimization subunit is further configured to: perform weighting processing on the set of bit-by-bit fluctuation entropies based on the Softmax function to obtain a set of bit-by-bit transformation entropy adjustment parameters, expressed by the formula:

[0062] a i =softmax(e i )

[0063] Where softmax represents the normalization exponential function, a i This represents the entropy adjustment parameter for the i-th bitwise transformation.

[0064] In other words, by introducing the Softmax function, the discrete entropy value is transformed into a bitwise transformation entropy adjustment parameter in the form of a probability distribution. Its exponential mechanism amplifies the weight difference between high-entropy components and low-entropy components, so that the inherent components that have a significant impact on device performance receive higher optimization weights and are thus given stronger attention in subsequent feature fusion.

[0065] More specifically, the dynamic optimization subunit is further configured to: based on the set of bit-by-bit transformation entropy adjustment parameters, fuse the set of device operating state timing feature vectors to obtain an optimized device operating state timing feature vector, expressed by the formula:

[0066]

[0067] Among them, v f This represents the timing feature vector of the optimized equipment operating status.

[0068] In other words, by using weighted processing, feature fusion no longer relies on fixed rules, but dynamically adjusts the contribution of each dimension according to the degree of information perturbation. This constructs a more discriminative and robust optimized equipment operating state time-series feature vector, significantly improving the information density and interpretability of the optimized equipment operating state time-series feature vector, and reducing the interference of noisy features on model decisions. Here, the process of performing differential entropy quantization dynamic optimization on the environmental data time-series feature vector to obtain the optimized environmental data time-series feature vector can also refer to the above-described processing process for the equipment operating state time-series feature vector.

[0069] Specifically, the feature saliency unit 142 is used to pass the device operating state-environment time-series response feature matrix through a feature saliency unit based on a spatial attention layer to obtain a saliency device operating state-environment time-series response feature matrix. It should be understood that each local spatial location in the device operating state-environment time-series response feature matrix represents different features; however, some of these features are important, while others are irrelevant. Therefore, in order to enhance the feature representation capability of the device operating state-environment time-series response feature matrix by utilizing the important feature information in the spatial domain, a feature saliency unit based on a spatial attention layer is further used to perform spatial dimension feature enhancement processing on the device operating state-environment time-series response feature matrix. The spatial attention layer can measure the importance of features at different local spatial locations in the device operating state-environment time-series response feature matrix, thereby achieving feature saliency. Specifically, the spatial attention layer calculates the attention weight of each local spatial location in the device operating state-environment time-series response feature matrix to represent the importance of the feature at that location, thereby highlighting important features and suppressing irrelevant features. Thus, the saliency-environment time-series response feature matrix obtained after processing by the spatial attention layer has a higher feature representation capability, which helps to improve the accuracy of subsequent performance evaluation of wind power generation equipment.

[0070] In a specific example of this application, the feature saliency unit 142 includes: a deep convolutional encoding subunit, used to perform deep convolutional encoding on the device operating state-environment temporal response feature matrix using the convolutional layer of the feature saliency unit to obtain an initial convolutional feature map; a spatial attention generation subunit, used to input the initial convolutional feature map into the spatial attention branch of the spatial attention mechanism model to obtain a spatial attention map; an activation subunit, used to apply the spatial attention map to the Softmax activation function to obtain a spatial attention feature map; a spatial attention application subunit, used to calculate the positional dot product of the spatial attention feature map and the initial convolutional feature map to obtain a spatial enhancement device operating state-environment temporal response feature map; and a pooling subunit, used to perform mean pooling along the channel dimension on the spatial enhancement device operating state-environment temporal response feature map to obtain the saliency device operating state-environment temporal response feature matrix.

[0071] In the aforementioned wind power equipment operation and maintenance management system 100, the equipment performance analysis module 150 is used to determine whether the performance of the wind power equipment is abnormal based on the salient equipment operating status-environment time-series response feature matrix. In a specific example of this application, the method for determining whether the performance of the wind power equipment is abnormal based on the salient equipment operating status-environment time-series response feature matrix is ​​to pass the salient equipment operating status-environment time-series response feature matrix through a classifier to obtain a classification result. The classification result is used to determine whether the performance of the wind power equipment is abnormal. It should be understood that a classifier is a machine learning model whose working principle is to map input features to corresponding category labels. In the technical solution of this application, the salient equipment operating status-environment time-series response feature matrix is ​​input into a classifier for classification operations. The classifier can learn the feature patterns in the salient equipment operating status-environment time-series response feature matrix, uncover the performance patterns of the wind power equipment, and associate the learned feature patterns with the performance standards of the wind power equipment, thereby determining whether the performance of the wind power equipment is abnormal, providing an important basis for the operation and maintenance of the wind power equipment. For example, when the classification results indicate that the equipment performance is abnormal, maintenance personnel can take appropriate repair measures in a timely manner to prevent production losses caused by equipment failure and provide support for the optimized operation and management of the equipment.

[0072] In summary, the operation and maintenance management system for wind power generation equipment according to the embodiments of this application is explained. It utilizes deep learning-based artificial intelligence technology to monitor and analyze the operational and environmental data of the wind power generation equipment, capturing the temporal variation characteristics of the wind power generation equipment's operational status and the temporal variation characteristics of the environmental data. Based on the temporal response characteristics of the wind power generation equipment's operational status relative to the environmental data, it intelligently determines whether there are any performance abnormalities in the equipment. This enables real-time monitoring, performance evaluation, and intelligent maintenance of the wind power generation equipment, thereby improving its operational efficiency and stability.

[0073] Figure 6 This is a flowchart illustrating a method for the operation, maintenance, and management of wind power generation equipment according to an embodiment of this application. Figure 6As shown, the operation and maintenance management method for wind power generation equipment according to an embodiment of this application includes the following steps: S110, acquiring operation data of the wind power generation equipment at multiple predetermined time points within a predetermined time period, and environmental data at multiple predetermined time points within the predetermined time period, wherein the operation data includes the equipment's power, rotor speed, current, and temperature, and the environmental data includes wind speed, wind direction, and air pressure; S120, performing time-series correlation encoding on the operation data at the multiple predetermined time points to obtain a time-series feature vector of the equipment's operation status; S130, extracting time-series features from the environmental data at the multiple predetermined time points to obtain a time-series feature vector of environmental data; S140, performing correlation encoding and correlation feature enhancement on the time-series feature vector of the equipment's operation status and the time-series feature vector of the environmental data to obtain a saliency-based equipment operation status-environment time-series response feature matrix; S150, determining whether there are any abnormalities in the performance of the wind power generation equipment based on the saliency-based equipment operation status-environment time-series response feature matrix.

[0074] Here, those skilled in the art will understand that the specific operations of each step in the above-described method for the operation, maintenance, and management of wind power equipment have been referenced above. Figures 1 to 5 The description of the operation and maintenance management system for wind power generation equipment is detailed here, and therefore, its repeated description will be omitted.

[0075] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. In the several embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for example, the module division is only a logical functional division, and other division methods may exist in actual implementation. The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0076] In addition, the functional modules in various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or hardware plus software functional modules.

[0077] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the present invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0078] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units stated in a system claim may also be implemented by a single unit through software or hardware.

[0079] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An operation and maintenance management system for wind power generation equipment, characterized in that, include: The wind power generation information acquisition module is used to acquire the operating data of the wind power generation equipment at multiple predetermined time points within a predetermined time period, as well as the environmental data at multiple predetermined time points within the predetermined time period. The operating data includes the power of the equipment, the wind turbine speed, the current and the temperature, and the environmental data includes the wind speed, the wind direction and the air pressure. The device operation status timing feature extraction module is used to perform timing association encoding on the operation data at the multiple predetermined time points to obtain the device operation status timing feature vector; An environmental temporal feature extraction module is used to extract temporal features from the environmental data at the multiple predetermined time points to obtain an environmental data temporal feature vector. The temporal correlation response module is used to perform correlation encoding and correlation feature enhancement on the temporal feature vector of the device operating state and the temporal feature vector of the environmental data to obtain a salient device operating state-environment temporal response feature matrix. The equipment performance analysis module is used to determine whether there are any abnormalities in the performance of the wind power generation equipment based on the salienced equipment operating status-environment time-series response feature matrix.

2. The operation and maintenance management system for wind power generation equipment according to claim 1, characterized in that, The device operating status time-series feature extraction module includes: The running data normalization unit is used to normalize the running data at the multiple predetermined time points to obtain the running data input matrix; The operating state timing feature extraction unit is used to input the operating data matrix into the device operating state timing association encoder based on a convolutional neural network to obtain the device operating state timing feature vector.

3. The operation and maintenance management system for wind power generation equipment according to claim 2, characterized in that, The data normalization unit is used for: The operational data at the multiple predetermined time points are arranged into the operational data input matrix according to the time dimension and the sample dimension.

4. The operation and maintenance management system for wind power generation equipment according to claim 3, characterized in that, The environmental temporal feature extraction module includes: An environmental data normalization unit is used to arrange the environmental data at the multiple predetermined time points into an environmental data input matrix according to the time dimension and the sample dimension. An environmental temporal feature extraction unit is used to extract the environmental data input matrix through an environmental temporal feature extractor based on a convolutional neural network to obtain the environmental data temporal feature vector.

5. The operation and maintenance management system for wind power generation equipment according to claim 4, characterized in that, The time-series correlation response module includes: The association unit is used to perform association encoding on the device operating state time-series feature vector and the environmental data time-series feature vector to obtain the device operating state-environment time-series response feature matrix; The feature saliency unit is used to pass the device operating state-environment time-series response feature matrix through a feature saliency unit based on a spatial attention layer to obtain the saliency device operating state-environment time-series response feature matrix.

6. The operation and maintenance management system for wind power generation equipment according to claim 5, characterized in that, The associated unit includes: The dynamic optimization subunit is used to perform differential entropy quantization dynamic optimization on the device operating state time-series feature vector and the environmental data time-series feature vector to obtain the optimized device operating state time-series feature vector and the optimized environmental data time-series feature vector. The feature fusion subunit performs weighted fusion of the optimized equipment operating state time-series feature vector and the optimized environment data time-series feature vector to obtain the equipment operating state-environment time-series response feature vector. The autocorrelation subunit is used to multiply the device operating state-environment time-series response feature vector with its own transpose to obtain the device operating state-environment time-series response feature matrix.

7. The operation and maintenance management system for wind power generation equipment according to claim 6, characterized in that, The dynamic optimization subunit is used for: Calculate the self-similarity matrix of the device operating state time-series feature vector, and perform key dimension reduction on the self-similarity matrix of the device operating state time-series feature vector to obtain a set of inherent component encoding vectors of the device operating state time-series feature vector; The set of encoded vectors of the inherent components of the device operating state temporal feature vector is input into a sequence encoder based on a forward LSTM model to obtain the set of encoded vectors of the context association of the inherent components of the device operating state temporal feature vector. Calculate the bitwise fluctuation entropy between the context-associative encoded vectors of the inherent components of the device operating state temporal feature vector and the encoded vectors of the inherent components of the device operating state temporal feature vector in the set of the device operating state temporal feature vectors and the encoded vectors of the inherent components of the device operating state temporal feature vector in the set of the device operating state temporal feature vectors to obtain the set of bitwise fluctuation entropy; The set of bit-by-bit fluctuation entropies is weighted using the Softmax function to obtain a set of bit-by-bit transformation entropy adjustment parameters; Based on the set of bit-by-bit transformation entropy adjustment parameters, the set of device operating state time-series feature vectors is fused to obtain an optimized device operating state time-series feature vector.

8. The operation and maintenance management system for wind power generation equipment according to claim 7, characterized in that, The feature saliency unit includes: A deep convolutional coding subunit is used to perform deep convolutional coding on the device operating state-environment time-series response feature matrix using the convolutional layer of the feature saliency to obtain an initial convolutional feature map; A spatial attention generation subunit is used to input the initial convolutional feature map into the spatial attention branch of the spatial attention mechanism model to obtain a spatial attention map; An activation subunit is used to pass the spatial attention map through a Softmax activation function to obtain a spatial attention feature map; The spatial attention application subunit is used to calculate the position-based dot product of the spatial attention feature map and the initial convolutional feature map to obtain the spatial augmentation device operating state-environment temporal response feature map. The pooling subunit is used to perform mean pooling along the channel dimension on the spatial augmentation device operating state-environment time-series response feature map to obtain the saliency device operating state-environment time-series response feature matrix.

9. The operation and maintenance management system for wind power generation equipment according to claim 8, characterized in that, The equipment performance analysis module is used for: The saliency of the equipment operating status-environment time-series response feature matrix is ​​passed through a classifier to obtain a classification result, which is used to determine whether the performance of the wind power generation equipment is abnormal.

10. A method for operation, maintenance and management of wind power generation equipment, characterized in that, include: The system acquires operational data of a wind power generation device at multiple predetermined time points within a predetermined time period, as well as environmental data at multiple predetermined time points within the predetermined time period. The operational data includes the device's power, rotor speed, current, and temperature, while the environmental data includes wind speed, wind direction, and air pressure. The operation data at the multiple predetermined time points are time-series correlated and encoded to obtain the time-series feature vector of the device operation status; Temporal feature extraction is performed on the environmental data at the multiple predetermined time points to obtain the environmental data temporal feature vector; The temporal feature vectors of the device operating status and the temporal feature vectors of the environmental data are correlated and encoded, and the correlation features are enhanced to obtain a salientized device operating status-environment temporal response feature matrix. Based on the saliency of the equipment operating status-environment time-series response feature matrix, it is determined whether there are any abnormalities in the performance of the wind power generation equipment.

Images