Offshore wind power equipment abnormal state monitoring method, device, equipment, medium and product

Abstract

The invention discloses an offshore wind power equipment abnormal state monitoring method and device, equipment, a medium and a product, and relates to the field of offshore wind power equipment abnormal state monitoring, and the method comprises the steps: inputting a sensor data set at each moment into a prediction model, and obtaining a target parameter prediction value at the ith moment; the prediction model is obtained by carrying out training on a CNN-LSTM-Attention model; the CNN-LSTM-Attention model comprises a one-dimensional convolutional neural network, a bidirectional long-short term memory network and an attention mechanism layer which are connected in sequence; calculating a loss value at the ith moment according to the target parameter predicted value and the target parameter true value at the ith moment; determining an exponentially weighted moving average control chart according to the loss value at the ith moment; and determining whether the offshore wind power equipment is in an abnormal state or not according to the loss value at the ith moment and the exponential weighted moving average control chart. According to the method, the abnormal state of the offshore wind power equipment can be efficiently and accurately monitored.

Description

Technical Field

[0001] The present application relates to the field of abnormal state monitoring of offshore wind power equipment, and in particular to a method, device, equipment, medium and product for abnormal state monitoring of offshore wind power equipment. Background Art

[0002] Offshore wind turbines face complex marine environments, such as extreme weather and corrosive seawater. These conditions lead to frequent failures, limited operational efficiency, and shortened equipment lifespans. Furthermore, their remote location from land leads to high maintenance costs, placing significant financial strain on wind farm operations. Consequently, efficient and accurate monitoring of abnormal conditions in offshore wind turbines has become crucial for ensuring stable wind farm operation and reducing maintenance costs. Summary of the Invention

[0003] The purpose of this application is to provide a method, device, equipment, medium and product for monitoring abnormal conditions of offshore wind power equipment, which can monitor abnormal conditions of offshore wind power equipment efficiently and accurately.

[0004] To achieve the above objectives, this application provides the following solutions:

[0005] In a first aspect, the present application provides a method for monitoring abnormal conditions of offshore wind power equipment, comprising:

[0006] A sensor data set of the offshore wind power equipment at each moment is obtained; the sensor data set includes data collected by all sensors on the offshore wind power equipment.

[0007] The sensor data set of the offshore wind turbine at each moment is input into the prediction model to obtain the target parameter prediction value of the offshore wind turbine at the i-th moment; the prediction model is obtained by training the CNN-LSTM-Attention model; the CNN-LSTM-Attention model includes a one-dimensional convolutional neural network, a bidirectional long short-term memory network, an attention mechanism layer and an output layer connected in sequence.

[0008] The loss value at the i-th moment is calculated according to the predicted value of the target parameter of the offshore wind power equipment at the i-th moment and the true value of the target parameter of the offshore wind power equipment at the i-th moment.

[0009] Determine the exponentially weighted moving average control chart based on the loss value at the i-th moment.

[0010] It is determined whether the offshore wind power equipment is in an abnormal state according to the loss value at the i-th moment and the exponentially weighted moving average control chart.

[0011] Optionally, the one-dimensional convolutional neural network specifically includes:

[0012] N one-dimensional convolutional layers and N pooling layers, the output end of the a-th one-dimensional convolutional layer is connected to the input end of the a-th pooling layer, and the output end of the pooling layer is connected to the input end of the bidirectional long short-term memory network, N is a positive integer greater than 1, 1≤a≤N.

[0013] Optionally, the bidirectional long short-term memory network specifically includes: a forward long short-term memory module, a reverse long short-term memory module and a splicing layer; the forward long short-term memory module and the reverse long short-term memory module each include N long short-term memory networks connected in sequence.

[0014] The output end of the ath pooling layer is connected to the input end of the ath long short-term memory network in the forward long short-term memory module and the input end of the ath long short-term memory network in the reverse long short-term memory module.

[0015] The output end of the last long short-term memory network in the forward long short-term memory module and the output end of the last long short-term memory network in the reverse long short-term memory module are both connected to the input end of the splicing layer, and the output end of the splicing layer is connected to the input end of the attention mechanism layer.

[0016] Optionally, determining an exponentially weighted moving average control chart according to the loss value at the i-th moment specifically includes:

[0017] Calculate the EWMA statistic at the i-th moment based on the loss value at the i-th moment and the EWMA statistic at the i-1-th moment.

[0018] The exponentially weighted moving average control chart is obtained based on the EWMA statistics at the i-th moment.

[0019] Optionally, determining whether the offshore wind power equipment is in an abnormal state according to the loss value at the i-th moment and the exponentially weighted moving average control chart specifically includes:

[0020] Determine the threshold value based on the exponentially weighted moving average control chart.

[0021] If the loss value at the i-th moment is greater than the threshold, it is determined that the offshore wind power equipment is in an abnormal state.

[0022] Optionally, the calculation formula for determining the threshold value based on the exponentially weighted moving average control chart is: Where L represents the threshold, μ represents the mean of the exponentially weighted moving average control chart, σ represents the variance of the exponentially weighted moving average control chart, and λ represents the weight coefficient.

[0023] In a second aspect, the present application provides an abnormal state monitoring device for offshore wind power equipment, comprising:

[0024] The acquisition module is used to acquire the sensor data set of the offshore wind power equipment at each moment; the sensor data set includes data collected by all sensors on the offshore wind power equipment.

[0025] The prediction module is used to input the sensor data set of the offshore wind turbine at each moment into the prediction model to obtain the predicted value of the target parameter of the offshore wind turbine at the i-th moment; the prediction model is obtained by training the CNN-LSTM-Attention model; the CNN-LSTM-Attention model includes a one-dimensional convolutional neural network, a bidirectional long short-term memory network, an attention mechanism and an output layer connected in sequence.

[0026] The loss value calculation module is used to calculate the loss value at the i-th moment based on the predicted value of the target parameter of the offshore wind power equipment at the i-th moment and the true value of the target parameter of the offshore wind power equipment at the i-th moment.

[0027] The exponentially weighted moving average control chart determination module is used to determine the exponentially weighted moving average control chart according to the loss value at the i-th moment.

[0028] The abnormal state judgment module is used to determine whether the offshore wind power equipment is in an abnormal state according to the loss value at the i-th moment and the exponentially weighted moving average control chart.

[0029] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the above-described methods for monitoring abnormal conditions of offshore wind power equipment.

[0030] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-mentioned methods for monitoring abnormal conditions of offshore wind power equipment.

[0031] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-mentioned methods for monitoring abnormal conditions of offshore wind power equipment.

[0032] According to the specific embodiments provided in this application, this application has the following technical effects:

[0033] The present application provides a method, device, equipment, medium and product for monitoring the abnormal state of offshore wind power equipment. The development of an Internet of Things online monitoring platform for offshore wind power equipment is an important means of maintaining offshore wind farms. Data is collected in real time through a sensor network and efficiently transmitted to the cloud background for analysis, thereby improving operation and maintenance efficiency and saving costs. The online monitoring system for offshore wind farms can centrally monitor the operation of offshore substations, converter stations, wind turbines, etc., and reflect the overall condition of wind turbines. However, existing systems mostly rely on traditional cross-border alarm modes, which have the problem of delayed response, or only provide fault warnings based on the time series of a single type of sensor data, ignoring the complex spatiotemporal correlation characteristics between data, making it difficult for the model to obtain the best monitoring accuracy. In summary, there is still room for optimization in the monitoring of abnormal states of offshore wind power equipment in terms of integrating the spatiotemporal correlation of multiple sensors, deepening data utilization efficiency, and improving monitoring accuracy. The CNN-LSTM-Attention provided in this application integrates 1D-CNN, Bi-LSTM and Attention mechanism. The data is first processed by 1D-CNN to capture spatial features, and then Bi-LSTM fuses the forward and backward time series information to extract temporal features. The Attention mechanism focuses on the key time series part, and is finally output by the output layer. The spatiotemporal characteristics of the data are extracted, and the spatiotemporal information of the data collected by the offshore wind power Internet of Things monitoring platform is integrated. The spatiotemporal correlation of different types of sensor data collected by the offshore wind power Internet of Things monitoring platform is effectively mined to achieve accurate prediction of abnormal conditions of offshore wind power equipment. Compared with the currently widely used model methods based on out-of-bounds alarms or simply relying on data time series relationships, this application improves the accuracy of equipment status prediction and can accurately monitor abnormal conditions of offshore wind power equipment. The exponentially weighted moving average control chart is used to efficiently identify early abnormal conditions of offshore wind power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0035] Figure 1 This is a diagram of an application environment of a method for monitoring abnormal conditions of offshore wind power equipment in one embodiment of the present application;

[0036] Figure 2 This is the structure diagram of the CNN-LSTM-Attention model;

[0037] Figure 3 A flow chart of a method for monitoring abnormal conditions of offshore wind power equipment provided in one embodiment of the present application;

[0038] Figure 4 A flowchart of a method for monitoring abnormal conditions of offshore wind power equipment provided in one embodiment of the present application;

[0039] Figure 5 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0041] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0042] The abnormal state monitoring method for offshore wind power equipment provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the sensor data set of the offshore wind power equipment at each moment to the server 104. After the server 104 receives the sensor data set of the offshore wind power equipment at each moment, the server 104 obtains the sensor data set of the offshore wind power equipment at each moment; the sensor data set includes data collected by all sensors on the offshore wind power equipment; the sensor data set of the offshore wind power equipment at each moment is input into the prediction model to obtain the target parameter prediction value of the offshore wind power equipment at the i-th moment; the prediction model is obtained by training the CNN-LSTM-Attention model; the CNN-LSTM-Attention model includes a one-dimensional convolutional neural network (1D-convolutional Neural Networks, 1D-CNN), a bidirectional long short-term memory network (Bidirectional-Long Short-Term Memory, BiLSTM), attention mechanism layer (Attention), and output layer; calculate the loss value at the i-th moment based on the predicted value of the target parameter of the offshore wind turbine at the i-th moment and the actual value of the target parameter of the offshore wind turbine at the i-th moment; determine an exponentially weighted moving average (EWMA) control chart based on the loss value at the i-th moment; and determine whether the offshore wind turbine is in an abnormal state based on the loss value at the i-th moment and the EWMA control chart. Server 104 can provide feedback on the obtained status to terminal 102. In addition, in some embodiments, the offshore wind turbine abnormal state monitoring method can also be implemented independently by server 104 or terminal 102. For example, terminal 102 can directly monitor the offshore wind turbine abnormal state based on the sensor data set of the offshore wind turbine at each moment, or server 104 can obtain the sensor data set of the offshore wind turbine at each moment from a data storage system and monitor the offshore wind turbine abnormal state based on the sensor data set of the offshore wind turbine at each moment. 1D-CNN is used to extract spatial features hidden in the data. Since there is a certain coupling relationship between the sensor data of offshore wind power equipment, the operating status of components at different locations generally reflects the overall operating status of the wind power equipment and can be regarded as a spatial feature. nIn the process, sensors installed at different locations of offshore wind turbines collect new data. The data collected by different sensors at the same time constitute a set of one-dimensional parameter sequences X = x1, x2, ..., x q . At each moment, the convolution kernel in 1D-CNN extracts spatial features from the input data. Since 1D-CNN only extracts the relationship between the monitoring variables at a specific moment, for time series data, 1D-CNN cannot fully capture the changes in features in the time series. Therefore, this application cascades a bidirectional long short-term memory network after 1D-CNN to complete the extraction and prediction regression of time features. The number of hidden units of the bidirectional long short-term memory network is 128. The role of the Attention layer is to improve the training efficiency of the network and enable the network to focus on the important parts of the time series data. Through the fusion of CNN and LSTM models, the model has stronger generalization ability and can more comprehensively capture the effective information in the data; the use of EWMA enables the model to quickly respond to anomalies in the equipment.

[0043] Terminal 102 may include, but is not limited to, various desktop computers, laptops, smartphones, tablet computers, IoT devices, and portable wearable devices. IoT devices may include smart speakers, smart TVs, smart air conditioners, and smart car devices. Portable wearable devices may include smart watches, smart bracelets, and head-mounted devices. Server 104 may be implemented as a standalone server or a server cluster consisting of multiple servers, or may be a cloud server.

[0044] In an exemplary embodiment, Figure 3 As shown, a method for monitoring abnormal conditions of offshore wind power equipment is provided. The method is executed by a computer device, specifically a computer device such as a terminal or a server, or a terminal and a server. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in FIG. 1 is used as an example to illustrate the method, which includes the following steps, wherein:

[0045] Step 201: Acquire a sensor data set of an offshore wind power device at each moment; the sensor data set includes data collected by all sensors on the offshore wind power device.

[0046] Step 202: Input the sensor data set of the offshore wind turbine at each moment into the prediction model to obtain the target parameter prediction value of the offshore wind turbine at the i-th moment; the prediction model is obtained by training the CNN-LSTM-Attention model; the CNN-LSTM-Attention model includes a one-dimensional convolutional neural network, a bidirectional long short-term memory network, an attention mechanism layer and an output layer connected in sequence.

[0047] Step 203: Calculate the loss value at the i-th moment based on the predicted value of the target parameter of the offshore wind turbine at the i-th moment and the actual value of the target parameter of the offshore wind turbine at the i-th moment. Specifically, the difference between the actual value of the target parameter of the offshore wind turbine at the i-th moment and the predicted value of the target parameter of the offshore wind turbine at the i-th moment is the loss value at the i-th moment.

[0048] Step 204: Determine an exponentially weighted moving average control chart based on the loss value at the i-th moment.

[0049] Step 205: Determine whether the offshore wind power equipment is in an abnormal state according to the loss value at the i-th moment and the exponentially weighted moving average control chart.

[0050] By implementing the above steps 201 to 205 , abnormal conditions of offshore wind power equipment can be monitored efficiently and accurately.

[0051] In another exemplary embodiment of the present application, Figure 4 As shown in the figure, the training of the CNN-LSTM-Attention model is as follows: obtaining samples, which include sensor data sets of the sample offshore wind turbine at each moment and the true values ​​of the target parameters at each moment; preprocessing the samples, and using the preprocessed samples to train the CNN-LSTM-Attention model with the goal of minimizing the loss function to obtain a prediction model. The loss function is: Among them, n is the total number of samples, y i and y' i is the true value at time i and the predicted value of the model.

[0052] In another exemplary embodiment of the present application, the sensor data set of the offshore wind turbine at each moment is input into the prediction model to obtain the target parameter prediction value of the offshore wind turbine at the i-th moment, specifically including:

[0053] The sensor data sets of offshore wind power equipment at each moment are preprocessed (1. Use linear interpolation to handle missing data values; 2. Remove meaningless data and wind curtailment data, including equipment failures, sensor abnormality records, and data with wind speeds exceeding the range of 4 m / s-25 m / s (select data with normal wind turbine operation status, and eliminate irrelevant interference data such as shutdown data and active power less than 0); 3. Normalize the data to eliminate the influence of different sensor data value ranges and dimensions).

[0054] The preprocessed sensor data set is input into the prediction model to obtain the target parameter prediction value of the offshore wind power equipment at the i-th moment.

[0055] In another exemplary embodiment of the present application, Figure 2 As shown, the one-dimensional convolutional neural network specifically includes:

[0056] N one-dimensional convolutional layers and N pooling layers, the output end of the a-th one-dimensional convolutional layer is connected to the input end of the a-th pooling layer, and the output end of the pooling layer is connected to the input end of the bidirectional long short-term memory network, N is a positive integer greater than 1, 1≤a≤N.

[0057] In another exemplary embodiment of the present application, a bidirectional long short-term memory network specifically includes: a forward long short-term memory module, a reverse long short-term memory module and a splicing layer; the forward long short-term memory module and the reverse long short-term memory module each include N long short-term memory networks connected in sequence.

[0058] The output end of the ath pooling layer is connected to the input end of the ath long short-term memory network in the forward long short-term memory module and the input end of the ath long short-term memory network in the reverse long short-term memory module.

[0059] The output end of the last long short-term memory network in the forward long short-term memory module and the output end of the last long short-term memory network in the reverse long short-term memory module are both connected to the input end of the splicing layer, and the output end of the splicing layer is connected to the input end of the attention mechanism layer.

[0060] In another exemplary embodiment of the present application, determining an exponentially weighted moving average control chart according to the loss value at the i-th moment specifically includes:

[0061] Calculate the EWMA statistic at the i-th moment based on the loss value at the i-th moment and the EWMA statistic at the i-1-th moment.

[0062] The exponentially weighted moving average control chart is obtained based on the EWMA statistics at the i-th moment.

[0063] In another exemplary embodiment of the present application, determining whether the offshore wind turbine is in an abnormal state according to the loss value at the i-th moment and the exponentially weighted moving average control chart specifically includes:

[0064] Determine the threshold value based on the exponentially weighted moving average control chart.

[0065] If the loss value at the i-th moment is greater than the threshold, it is determined that the offshore wind power equipment is in an abnormal state.

[0066] In another exemplary embodiment of the present application, the EWMA statistic (weighted average of sample means) at the i-th moment is calculated based on the loss value at the i-th moment and the EWMA statistic at the i-1-th moment, specifically: According to the formula Z i =λR i +(1-λ)Z i-1 Calculate the EWMA statistic Z at the i-th momenti , where R i Represents the loss value at the i-th moment, Z i-1 represents the EWMA statistic at the i-1th moment, λ is the weight coefficient, and this application takes 0.2.

[0067] The vertical axis of the exponentially weighted moving average control chart is Z i , the horizontal axis is time i, and the center line is Z0 (set to the mean value of the MSE of the equipment during normal operation). In another exemplary embodiment of the present application, the calculation formula for determining the threshold value according to the exponentially weighted moving average control chart is: Where L represents the threshold, μ represents the mean of the exponentially weighted moving average control chart, σ represents the variance of the exponentially weighted moving average control chart, and λ represents the weight coefficient.

[0068] In another exemplary embodiment of the present application, the above-mentioned method for monitoring abnormal conditions of offshore wind power equipment is explained by taking the offshore wind power equipment as a transformer as an example. The sensor data set includes transformer core grounding current, atmospheric corrosion detection, micrometeorological sensor and SF6 digital density meter, and the target parameter is transformer core temperature.

[0069] Step 1: Select data such as transformer core grounding current, atmospheric corrosion detection, micrometeorological sensors, and SF6 digital density meters as model inputs. Preprocess the collected data. Select data from wind turbines operating normally and exclude irrelevant interference data such as shutdown data and data with active power less than 0.

[0070] Step 2: Build a prediction model. The preprocessed data is first passed through a single-layer 1D-CNN with a convolution kernel size of 5 and a total of 32 kernels. The resulting feature vector undergoes a pooling layer for dimensionality reduction and is then input into a bidirectional long short-term memory network with 128 neurons. The output of the bidirectional long short-term memory network then passes through an attention layer and a fully connected layer (output layer) for output.

[0071] Step 3: The model uses the Mean Square Error (MSE) to predict the error between the actual value and the predicted value. It also uses the EWMA to determine whether the collected equipment data deviates from the predicted results, thereby determining whether the equipment's health status is abnormal. When the MSE value exceeds the EWMA threshold, it indicates that the operating data of the offshore wind turbine equipment deviates from the model's prediction results, thus determining that the equipment's health status is abnormal.

[0072] Based on the same inventive concept, embodiments of the present application also provide an offshore wind turbine abnormal state monitoring device for implementing the aforementioned offshore wind turbine abnormal state monitoring method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the offshore wind turbine abnormal state monitoring device provided below can be found in the aforementioned limitations of the offshore wind turbine abnormal state monitoring method and will not be further elaborated here.

[0073] In an exemplary embodiment, a device for monitoring abnormal conditions of offshore wind power equipment is provided, comprising:

[0074] The acquisition module is used to acquire the sensor data set of the offshore wind power equipment at each moment; the sensor data set includes data collected by all sensors on the offshore wind power equipment.

[0075] The prediction module is used to input the sensor data set of the offshore wind turbine at each moment into the prediction model to obtain the predicted value of the target parameter of the offshore wind turbine at the i-th moment; the prediction model is obtained by training the CNN-LSTM-Attention model; the CNN-LSTM-Attention model includes a one-dimensional convolutional neural network, a bidirectional long short-term memory network, an attention mechanism and an output layer connected in sequence.

[0076] The loss value calculation module is used to calculate the loss value at the i-th moment based on the predicted value of the target parameter of the offshore wind power equipment at the i-th moment and the true value of the target parameter of the offshore wind power equipment at the i-th moment.

[0077] The exponentially weighted moving average control chart determination module is used to determine the exponentially weighted moving average control chart according to the loss value at the i-th moment.

[0078] The abnormal state judgment module is used to determine whether the offshore wind power equipment is in an abnormal state according to the loss value at the i-th moment and the exponentially weighted moving average control chart.

[0079] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 5As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store abnormal state monitoring data of offshore wind power equipment. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for monitoring the abnormal state of offshore wind power equipment is implemented.

[0080] Those skilled in the art will understand that Figure 5 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application and does not constitute a limitation on the computer device to which the solution of the present application is applied. A specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the above-mentioned method embodiments when executing the computer program.

[0081] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the above-mentioned method embodiments when executed by a processor.

[0082] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the above method embodiments are implemented.

[0083] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0084] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0085] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0086] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0087] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for monitoring abnormal conditions of offshore wind power equipment, characterized in that: The method for monitoring abnormal conditions of offshore wind power equipment comprises: Acquire a sensor data set of the offshore wind power equipment at each moment; the sensor data set includes data collected by all sensors on the offshore wind power equipment; Inputting sensor data sets of the offshore wind turbine at each moment into a prediction model to obtain a predicted value of a target parameter of the offshore wind turbine at the i-th moment; the prediction model is obtained by training a CNN-LSTM-Attention model; the CNN-LSTM-Attention model comprises a one-dimensional convolutional neural network, a bidirectional long short-term memory network, an attention mechanism layer, and an output layer connected in sequence; Calculate the loss value at the i-th moment according to the predicted value of the target parameter of the offshore wind power equipment at the i-th moment and the actual value of the target parameter of the offshore wind power equipment at the i-th moment; Determine the exponentially weighted moving average control chart based on the loss value at the i-th moment; It is determined whether the offshore wind power equipment is in an abnormal state according to the loss value at the i-th moment and the exponentially weighted moving average control chart.

2. The method for monitoring abnormal conditions of offshore wind power equipment according to claim 1, characterized in that: The one-dimensional convolutional neural network specifically includes: N one-dimensional convolutional layers and N pooling layers, the output end of the a-th one-dimensional convolutional layer is connected to the input end of the a-th pooling layer, and the output end of the pooling layer is connected to the input end of the bidirectional long short-term memory network, N is a positive integer greater than 1, 1≤a≤N.

3. The method for monitoring abnormal conditions of offshore wind power equipment according to claim 2, characterized in that: A bidirectional long short-term memory network, specifically comprising: a forward long short-term memory module, a reverse long short-term memory module, and a splicing layer; the forward long short-term memory module and the reverse long short-term memory module each comprise N sequentially connected long short-term memory networks; The output end of the a-th pooling layer is connected to the input end of the a-th long short-term memory network in the forward long short-term memory module and the input end of the a-th long short-term memory network in the reverse long short-term memory module respectively; The output end of the last long short-term memory network in the forward long short-term memory module and the output end of the last long short-term memory network in the reverse long short-term memory module are both connected to the input end of the splicing layer, and the output end of the splicing layer is connected to the input end of the attention mechanism layer.

4. The method for monitoring abnormal conditions of offshore wind power equipment according to claim 1, characterized in that: Determine the exponentially weighted moving average control chart based on the loss value at the i-th moment, specifically including: Calculate the EWMA statistic at the i-th moment based on the loss value at the i-th moment and the EWMA statistic at the i-1-th moment; The exponentially weighted moving average control chart is obtained based on the EWMA statistics at the i-th moment.

5. The method for monitoring abnormal conditions of offshore wind power equipment according to claim 1, characterized in that: Determining whether the offshore wind power equipment is in an abnormal state according to the loss value at the i-th moment and the exponentially weighted moving average control chart specifically includes: Determine thresholds based on an exponentially weighted moving average control chart; If the loss value at the i-th moment is greater than the threshold, it is determined that the offshore wind power equipment is in an abnormal state.

6. The method for monitoring abnormal conditions of offshore wind power equipment according to claim 5, characterized in that: The calculation formula for determining the threshold value based on the exponentially weighted moving average control chart is: Where L represents the threshold, μ represents the mean of the exponentially weighted moving average control chart, σ represents the variance of the exponentially weighted moving average control chart, and λ represents the weight coefficient.

7. A device for monitoring abnormal conditions of offshore wind power equipment, characterized in that: The offshore wind power equipment abnormal state monitoring device comprises: An acquisition module is used to acquire a sensor data set of the offshore wind power equipment at each moment; the sensor data set includes data collected by all sensors on the offshore wind power equipment; A prediction module is configured to input sensor data sets of the offshore wind turbine at each moment into a prediction model to obtain a predicted value of a target parameter of the offshore wind turbine at the i-th moment; the prediction model is obtained by training a CNN-LSTM-Attention model; the CNN-LSTM-Attention model comprises a one-dimensional convolutional neural network, a bidirectional long short-term memory network, an attention mechanism, and an output layer connected in sequence; A loss value calculation module is used to calculate the loss value at the i-th moment based on the predicted value of the target parameter of the offshore wind power equipment at the i-th moment and the actual value of the target parameter of the offshore wind power equipment at the i-th moment; An exponentially weighted moving average control chart determination module, used to determine an exponentially weighted moving average control chart according to the loss value at the i-th moment; The abnormal state judgment module is used to determine whether the offshore wind power equipment is in an abnormal state according to the loss value at the i-th moment and the exponentially weighted moving average control chart.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for monitoring abnormal conditions of offshore wind power equipment according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for monitoring abnormal conditions of offshore wind power equipment according to any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for monitoring abnormal conditions of offshore wind power equipment according to any one of claims 1 to 6 is implemented.

Images