Power generation equipment abnormality monitoring method, device, storage medium and electronic equipment

Abstract

The present application discloses a method, device, storage medium and electronic device for monitoring abnormalities of power generation equipment, which relates to the field of artificial intelligence technology, specifically to the fields of deep learning, neural networks, and natural language processing, and can be applied to scenarios such as new energy, carbon neutrality, and carbon peak. The specific implementation scheme is: by acquiring detection text data, preprocessing the detection text data, reconstructing the detection text data in reverse order based on preset weights to obtain reconstructed text data, calculating the maximum likelihood estimate based on the detection text data and the reconstructed text data, obtaining the abnormal probability of the power generation equipment, and determining the abnormal monitoring results of the power generation equipment based on the abnormal probability of the power generation equipment. The present application realizes the prediction of abnormal conditions of power generation equipment and improves the accuracy of abnormal monitoring of power generation equipment.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, specifically to the fields of deep learning, neural networks, and natural language processing, and can be applied to scenarios such as new energy, carbon neutrality, and carbon peak. In particular, it relates to a method, device, storage medium, and electronic device for abnormal monitoring of power generation equipment. Background Art

[0002] In the related art, power generation equipment is installed outdoors and is prone to abnormal operation under the conditions of severe cold, wind, rain and other natural environments. In addition, the current abnormal monitoring of power generation equipment basically relies on sensors, but once the sensors have a large amount of data accumulated, they are prone to failure. Therefore, how to monitor the abnormal conditions of equipment has become an urgent problem to be solved. Summary of the invention

[0003] The present application provides a method, device, storage medium and electronic device for monitoring abnormality of power generation equipment.

[0004] According to a first aspect of the present application, a method for monitoring abnormality of power generation equipment is provided, comprising:

[0005] Get the detected text data;

[0006] Preprocessing the detected text data;

[0007] Reconstructing the detected text data in reverse order based on preset weights to obtain reconstructed text data;

[0008] Calculating a maximum likelihood estimation value based on the detection text data and the reconstruction text data to obtain an abnormal probability of the power generation equipment;

[0009] According to the abnormal probability of the power generation equipment, the abnormal monitoring result of the power generation equipment is determined.

[0010] According to an embodiment of the present disclosure, the step of reconstructing the detected text data in reverse order based on preset weights to obtain reconstructed text data includes:

[0011] Inputting the preprocessed detection text data into a pre-trained reconstruction model; the reconstruction model includes a conversion module, a feature extraction module and a reverse reconstruction module;

[0012] Acquire text vector data obtained by converting the detected text data by a conversion module;

[0013] Acquiring text feature information obtained by a feature extraction module performing feature extraction on the text vector data based on the preset weight;

[0014] The reconstructed text data is obtained by the reverse reconstruction module performing reverse reconstruction on the text feature information based on the preset weight.

[0015] According to one embodiment of the present disclosure, the reconstruction model is pre-trained by the following steps:

[0016] Get the detection text sample data;

[0017] Preprocessing the detection text sample data;

[0018] Inputting the preprocessed detection text sample data into a target neural network model; the target neural network model includes a conversion module, a feature extraction module and a reverse reconstruction module;

[0019] Inputting the preprocessed detection text sample data into the conversion module to obtain text vector sample data;

[0020] Inputting the text vector sample data into the feature extraction module to obtain text feature sample information based on the preset weight;

[0021] The text feature sample information is input into the reverse reconstruction module to obtain reconstructed text data that is reversely reconstructed based on the preset weights.

[0022] According to an embodiment of the present disclosure, the reconstruction model further includes a self-attention module; wherein the inputting of the text vector sample data into the feature extraction module to obtain the text feature sample information based on the preset weights includes:

[0023] Inputting the text vector sample data into a self-attention module to obtain weight information of the text vector sample data;

[0024] The text vector sample data and the weight information are input into the feature extraction module to obtain text feature sample information based on the preset weight.

[0025] According to an embodiment of the present disclosure, the step of inputting the text feature sample information into the reverse reconstruction module to obtain reconstructed text data that is reversely reconstructed based on the preset weights includes:

[0026] Inputting the text feature sample information into a self-attention module to obtain weight information of the text feature sample information;

[0027] The text feature sample information and the weight information are input into the reverse reconstruction module to obtain reconstructed text data based on the preset weights.

[0028] According to an embodiment of the present disclosure, determining the abnormality monitoring result of the power generation equipment according to the abnormality probability of the power generation equipment includes:

[0029] Obtain the preset data anomaly probability threshold;

[0030] In response to the abnormal probability of the power generation equipment being greater than the preset data abnormal probability threshold, determining that the abnormal monitoring result of the power generation equipment is abnormal;

[0031] In response to the abnormal probability of the power generation equipment being less than or equal to the preset data abnormal probability threshold, it is determined that the abnormal monitoring result of the power generation equipment is normal.

[0032] According to a second aspect of the present application, a power generation equipment abnormality monitoring device is provided, comprising:

[0033] An acquisition module is used to acquire detection text data;

[0034] A preprocessing module, used for preprocessing the detected text data;

[0035] A reconstruction module, used for reconstructing the detected text data in reverse order based on preset weights to obtain reconstructed text data;

[0036] A calculation module, used for calculating a maximum likelihood estimation value according to the detection text data and the reconstructed text data to obtain an abnormal probability of the power generation equipment;

[0037] The determination module is used to determine the abnormality monitoring result of the power generation equipment according to the abnormality probability of the power generation equipment.

[0038] According to one embodiment of the present disclosure, the reconstruction module includes:

[0039] An input submodule, used for inputting the pre-processed detection text data into a pre-trained reconstruction model; the reconstruction model includes a conversion module, a feature extraction module and a reverse reconstruction module;

[0040] A first acquisition submodule, used for acquiring text vector data obtained by converting the detected text data by a conversion module;

[0041] A second acquisition submodule is used to acquire text feature information obtained by a feature extraction module through feature extraction of the text vector data based on the preset weight;

[0042] The third acquisition submodule is used to obtain the reconstructed text data obtained by the reverse reconstruction module by reversely reconstructing the text feature information based on the preset weight.

[0043] According to one embodiment of the present disclosure, the device further includes:

[0044] A pre-training module is used to pre-train the reconstruction model; wherein the pre-training module is specifically used to:

[0045] Get the detection text sample data;

[0046] Preprocessing the detection text sample data;

[0047] Inputting the preprocessed detection text sample data into a target neural network model; the target neural network model includes a conversion module, a feature extraction module and a reverse reconstruction module;

[0048] Inputting the preprocessed detection text sample data into the conversion module to obtain text vector sample data;

[0049] Inputting the text vector sample data into the feature extraction module to obtain text feature sample information;

[0050] The text feature sample information is input into the reverse reconstruction module to obtain reconstructed text data that is reversely reconstructed based on the preset weights.

[0051] According to an embodiment of the present disclosure, the reconstruction model further includes a self-attention module; wherein the pre-training module is specifically used for:

[0052] Inputting the text vector sample data into a self-attention module to obtain weight information of the text vector sample data;

[0053] The text vector sample data and the weight information are input into the feature extraction module to obtain text feature sample information based on the preset weight.

[0054] According to one embodiment of the present disclosure, the pre-training module is specifically used for:

[0055] Inputting the text feature sample information into a self-attention module to obtain weight information of the text feature sample information;

[0056] The text feature sample information and the weight information are input into the reverse reconstruction module to obtain reconstructed text data based on the preset weights.

[0057] According to one embodiment of the present disclosure, the determination module includes:

[0058] The fourth acquisition submodule is used to obtain a preset data abnormality probability threshold;

[0059] A first determination submodule, configured to determine that the abnormality monitoring result of the power generation equipment is abnormal in response to the abnormality probability of the power generation equipment being greater than the preset data abnormality probability threshold;

[0060] The second determination submodule is configured to determine that the abnormality monitoring result of the power generation equipment is normal in response to the abnormality probability of the power generation equipment being less than or equal to the preset data abnormality probability threshold.

[0061] According to a third aspect of the present application, an electronic device is provided, characterized in that it includes:

[0062] at least one processor; and

[0063] a memory communicatively connected to the at least one processor; wherein,

[0064] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in the first aspect.

[0065] According to a fourth aspect of the present application, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method described in the first aspect.

[0066] According to a fifth aspect of the present application, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the steps of the method for monitoring abnormalities of power generation equipment as described in the first aspect are implemented.

[0067] According to the technical solution of the present application, by reconstructing the detection text data in reverse order based on preset weights to obtain reconstructed text data, and calculating the maximum likelihood estimate based on the detection text data and the reconstructed text data to obtain the abnormal probability of the power generation equipment, the abnormal situation of the power generation equipment can be predicted, thereby improving the accuracy of abnormal monitoring of the power generation equipment.

[0068] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present application.

[0070] Figure 1 is a schematic diagram according to the first embodiment of the present application;

[0071] Figure 2 is a schematic diagram according to a second embodiment of the present application;

[0072] Figure 3 is a schematic diagram according to a third embodiment of the present application;

[0073] Figure 4is a schematic diagram according to a fourth embodiment of the present application;

[0074] Figure 5 It is a block diagram of an electronic device used to implement the power generation equipment abnormality monitoring method of an embodiment of the present application. DETAILED DESCRIPTION

[0075] The following is a description of exemplary embodiments of the present application in conjunction with the accompanying drawings, including various details of the embodiments of the present application to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0076] It should be noted that in the related art, power generation equipment is installed outdoors and is prone to abnormal operation under natural conditions such as severe cold, wind and rain. In addition, the current abnormal monitoring of power generation equipment basically relies on sensors, but once the sensors have a large amount of data accumulated, they are prone to failure. Therefore, how to monitor the abnormal conditions of equipment has become an urgent problem to be solved.

[0077] Based on the above problems, the present application proposes a method, device, storage medium and electronic device for monitoring abnormalities of power generation equipment. The detection text data is reversely reconstructed based on preset weights to obtain reconstructed text data. The maximum likelihood estimate is calculated based on the detection text data and the reconstructed text data to obtain the abnormal probability of the power generation equipment. This can realize the prediction of abnormal conditions of the power generation equipment and improve the accuracy of abnormal monitoring of the power generation equipment.

[0078] Figure 1 It is a schematic diagram according to the first embodiment of the present application. It should be noted that the power generation equipment abnormality monitoring method in the embodiment of the present application can be used for the power generation equipment abnormality monitoring device in the embodiment of the present application, and the device can be configured in a computer device. Figure 1 As shown, the power generation equipment abnormality monitoring method includes the following steps:

[0079] Step 101, obtaining detection text data.

[0080] It should be noted that the detection text data can be obtained by monitoring the power generation equipment with sensors, manually inputting monitoring parameters of the power generation equipment, monitoring the power generation equipment with smart meters, etc. As a possible example, the data obtained by the smart meter monitoring the power generation equipment can be text data or image data. For the image data, it can be converted into text data by methods such as optical character recognition (OCR).

[0081] Step 102: pre-process the detected text data.

[0082] It is understandable that the acquired detection text data contains some unnecessary irrelevant data, so it is necessary to clean the data through preprocessing to ensure data quality.

[0083] Step 103: Reconstruct the detected text data in reverse order based on preset weights to obtain reconstructed text data.

[0084] As a possible example, the detected text data is reconstructed in reverse order according to preset weights to obtain reconstructed text data based on the preset weights.

[0085] It should be noted that the above-mentioned preset weights may be preset weight values ​​for different text features according to actual needs.

[0086] Step 104, calculating the maximum likelihood estimation value based on the detected text data and the reconstructed text data to obtain the abnormal probability of the power generation equipment.

[0087] As a possible example, the abnormal probability of the error between the detection text data and the reconstructed text data is estimated by using the maximum likelihood method, thereby obtaining the abnormal probability of the power generation equipment.

[0088] Step 105, determining the abnormal monitoring result of the power generation equipment according to the abnormal probability of the power generation equipment.

[0089] As a possible example, in response to the abnormal probability of the power generation equipment reaching a preset requirement, it can be determined that the power generation equipment has an abnormal situation, and in response to the abnormal probability of the power generation equipment not reaching the preset requirement, it can be determined that the power generation equipment has no abnormal situation.

[0090] According to the power generation equipment abnormality monitoring method of the embodiment of the present application, by acquiring detection text data, preprocessing the detection text data, and reconstructing the detection text data in reverse order based on preset weights to obtain reconstructed text data, a maximum likelihood estimate is calculated based on the detection text data and the reconstructed text data to obtain the power generation equipment abnormality probability, and based on the power generation equipment abnormality probability, the power generation equipment abnormality monitoring result is determined, thereby being able to accurately predict the power generation equipment abnormality and improving the accuracy of power generation equipment abnormality monitoring.

[0091] In order to ensure that it is easy to judge whether the operating status of the power generation equipment is due to factors of the equipment itself or transmission problems, etc., optionally, detection text sample data is obtained to preprocess the detection text sample data, the preprocessed detection text sample data is input into the target neural network model, the preprocessed detection text sample data is input into the conversion module to obtain text vector sample data, and the text feature sample information is input into the reverse reconstruction module to obtain reconstructed text data that is reversely reconstructed based on preset weights. Figure 2 It is a schematic diagram according to the second embodiment of the present application. It should be noted that the power generation equipment abnormality monitoring method of the present application embodiment can be executed by the power generation equipment abnormality monitoring device in the present application embodiment. In some embodiments of the present application, such as Figure 2 As shown, the power generation equipment abnormality monitoring method includes the following steps:

[0092] Step 210, obtaining detection text data.

[0093] In the embodiments of the present application, step 210 may be implemented in any of the embodiments of the present application, and the embodiments of the present application do not limit this and will not be described in detail.

[0094] Step 220, preprocessing the detected text data.

[0095] In the embodiments of the present application, step 220 may be implemented in any of the embodiments of the present application, and the embodiments of the present application do not limit this and will not be described in detail.

[0096] Step 230, input the preprocessed detection text data into a pre-trained reconstruction model; the reconstruction model includes a conversion module, a feature extraction module and a reverse reconstruction module.

[0097] In this embodiment of the present application, the reconstruction model is pre-trained by the following steps:

[0098] Step 231, obtaining detection text sample data.

[0099] It should be noted that the detection text sample data can be obtained by monitoring the power generation equipment with sensors, manually inputting the monitoring parameters of the power generation equipment, monitoring the power generation equipment with smart meters, etc., wherein, as a possible example, the data obtained by monitoring the power generation equipment with smart meters can be text data or image data. For image data, text data can be obtained by methods such as optical character recognition (OCR).

[0100] Step 232, pre-processing the detected text sample data.

[0101] It is understandable that the acquired detection text sample data contains some unnecessary irrelevant data, so it is necessary to clean the data through preprocessing to ensure data quality.

[0102] Step 233, input the preprocessed detection text sample data into the target neural network model.

[0103] Among them, in the embodiment of the present application, the target neural network model includes a conversion module, a feature extraction module and a reverse reconstruction module.

[0104] As a possible example, the target neural network model may be a long short-term memory (LSTM) model, and the conversion module may be a Transformer conversion module.

[0105] Step 234, input the preprocessed detection text sample data into the conversion module to obtain text vector sample data.

[0106] As a possible example, the preprocessed detection text sample data is input into a conversion module, and the conversion module converts the preprocessed detection text sample data into text vector sample data.

[0107] Step 235 , input the text vector sample data into a feature extraction module to obtain text feature sample information based on preset weights.

[0108] As a possible example, the text vector sample data is input into a feature extraction module, and the feature extraction module performs feature extraction on the text vector sample data based on preset weights to obtain text feature sample information.

[0109] In the embodiment of the present application, the text vector sample data is input into the feature extraction module to obtain text feature sample information based on preset weights, including:

[0110] Step 2351, input the text vector sample data into the self-attention module to obtain the weight information of the text vector sample data.

[0111] It can be understood that the text vector sample data is input into the self-attention module, and the self-attention module obtains the weight information of the text vector sample data according to the above preset weights.

[0112] As a possible example, multiple text vector sample data are input into the self-attention module, and the preset weight can be a weight matrix. The self-attention module can calculate the weight information of each of the multiple text vector sample data according to the weight matrix.

[0113] Step 2352, input the text vector sample data and weight information into the feature extraction module to obtain text feature sample information based on preset weights.

[0114] It should be noted that the feature extraction module of LSTM performs feature extraction on the text vector sample data according to the weight information calculated by the self-attention module to obtain the text feature sample information based on the preset weights, that is, the target text feature sample information

[0115] Step 236: input the text feature sample information into the reverse reconstruction module to obtain reconstructed text data that is reverse reconstructed based on preset weights.

[0116] In this embodiment of the present application, the reconstruction model also includes a self-attention module.

[0117] In the embodiment of the present application, the text feature sample information is input into the reverse reconstruction module to obtain the reconstructed text data that is reversely reconstructed based on the preset weights, including:

[0118] Step 2361, input the text feature sample information into the self-attention module to obtain the weight information of the text feature sample information.

[0119] It can be understood that the text feature sample information is input into the self-attention module, and the self-attention module obtains the weight information of the text feature sample information according to the above preset weights.

[0120] As a possible example, multiple text feature sample information is input into the self-attention module, the preset weight may be a weight matrix, and the self-attention module may calculate the weight information of each of the multiple text feature sample information according to the weight matrix.

[0121] Step 2362, input the text feature sample information and weight information into the reverse reconstruction module to obtain reconstructed text data.

[0122] It should be noted that the LSTM reverse reconstruction module performs reverse reconstruction of the text feature sample information based on the weight information calculated by the self-attention module to obtain reconstructed text data based on preset weights, namely, the target reconstructed text data. By reconstructing the text feature sample information in reverse order, the text feature sample information is effectively enhanced, thereby improving the reconstruction ability of the reconstruction model, and thus being able to more accurately monitor abnormal conditions of power generation equipment.

[0123] As a possible example, the above reverse reconstruction may be to arrange the text feature sample matrix in reverse order, and multiply the text feature sample matrix after the reverse order arrangement by the above weight matrix to obtain the reconstructed text data.

[0124] Step 240, calculating the maximum likelihood estimation value based on the detected text data and the reconstructed text data to obtain the abnormal probability of the power generation equipment.

[0125] In the embodiments of the present application, step 240 may be implemented in any of the embodiments of the present application, and the embodiments of the present application do not limit this and will not be described in detail.

[0126] Step 250, determining the abnormal monitoring result of the power generation equipment according to the abnormal probability of the power generation equipment.

[0127] In the embodiments of the present application, step 250 may be implemented in any of the embodiments of the present application, and the embodiments of the present application do not limit this and will not be described in detail.

[0128] According to the abnormality monitoring method for power generation equipment in an embodiment of the present application, detection text sample data is obtained to pre-process the detection text sample data, the pre-processed detection text sample data is input into a target neural network model, the pre-processed detection text sample data is input into a conversion module to obtain text vector sample data, and the text feature sample information is input into a reverse reconstruction module to obtain reconstructed text data that is reversely reconstructed based on preset weights, thereby facilitating the judgment of whether the operating status of the power generation equipment is a factor of the equipment itself or a transmission problem, etc.

[0129] In order to ensure the accuracy of the judgment of the abnormal monitoring results of the power generation equipment, optionally, a preset data abnormality probability threshold is obtained, and in response to the abnormal probability of the power generation equipment being greater than the preset data abnormality probability threshold, the abnormal monitoring result of the power generation equipment is determined to be abnormal; in response to the abnormal probability of the power generation equipment being less than or equal to the preset data abnormality probability threshold, the abnormal monitoring result of the power generation equipment is determined to be normal. Figure 3 It is a schematic diagram according to the third embodiment of the present application. It should be noted that the power generation equipment abnormality monitoring method of the embodiment of the present application can be executed by the power generation equipment abnormality monitoring device in the embodiment of the present application. In some embodiments of the present application, such as Figure 3 As shown, the power generation equipment abnormality monitoring method includes the following steps:

[0130] Step 301, obtaining detection text data.

[0131] Step 302: pre-process the detected text data.

[0132] In the embodiment of the present application, the above preprocessing may include at least one or more of the following items 1)-4):

[0133] 1) De-noising the detected text data;

[0134] 2) Perform sentence completion on the detected text data;

[0135] 3) Unify the format of the detected text data;

[0136] 4) Perform normalized data processing on the detected text data.

[0137] Step 303: Reconstruct the detected text data in reverse order based on preset weights to obtain reconstructed text data.

[0138] It should be noted that there are multiple detection text data as mentioned above, and the multiple detection text data are reconstructed in reverse order to obtain reconstructed text data corresponding to each of the multiple detection text data.

[0139] Step 304, calculating the maximum likelihood estimation value based on the detected text data and the reconstructed text data to obtain the abnormal probability of the power generation equipment.

[0140] It should be noted that the maximum likelihood estimation values ​​are calculated respectively based on a plurality of detected text data and the corresponding reconstructed text data.

[0141] Step 305: Obtain a preset data anomaly probability threshold.

[0142] It should be noted that the above data anomaly probability threshold is a preset threshold, and the above data anomaly probability threshold may be a specific probability value range.

[0143] Step 306 , in response to the abnormal probability of the power generation equipment being greater than a preset data abnormal probability threshold, determining that the abnormal monitoring result of the power generation equipment is abnormal.

[0144] It can be understood that in response to the abnormal probability of the power generation equipment being greater than the preset data abnormal probability threshold, the detection text data corresponding to the abnormal probability of the power generation equipment can be determined to be abnormal, and the abnormal monitoring result of the power generation equipment of the detection text data is determined to be abnormal.

[0145] Step 307 : In response to the abnormal probability of the power generation equipment being less than or equal to a preset data abnormal probability threshold, determining that the abnormal monitoring result of the power generation equipment is normal.

[0146] It can be understood that in response to the abnormal probability of the power generation equipment being less than or equal to the preset data abnormal probability threshold, it can be determined that the detection text data corresponding to the abnormal probability of the power generation equipment is normal, and the abnormal monitoring result of the power generation equipment of the detection text data is normal.

[0147] As an example of a possible implementation, after obtaining the abnormal monitoring results of the power generation equipment, the abnormal monitoring results of the power generation equipment can be output and displayed in the following ways: text display; voice broadcast; outbound call terminal; SMS delivery; email delivery; smart speaker; voice wake-up; large-screen display.

[0148] According to the abnormality monitoring method for power generation equipment in an embodiment of the present application, a preset data abnormality probability threshold is obtained, and in response to the abnormality probability of the power generation equipment being greater than the preset data abnormality probability threshold, the abnormality monitoring result of the power generation equipment is determined to be abnormal, and in response to the abnormality probability of the power generation equipment being less than or equal to the preset data abnormality probability threshold, the abnormality monitoring result of the power generation equipment is determined to be normal, thereby improving the accuracy of the judgment of the abnormality monitoring result of the power generation equipment.

[0149] In order to implement the above-mentioned embodiment, the present application proposes a device for monitoring abnormality of power generation equipment.

[0150] Figure 4 is a schematic diagram according to the fourth embodiment of the present application. Figure 4 As shown, the device comprises:

[0151] The acquisition module 401 is used to acquire the detected text data.

[0152] The preprocessing module 402 is used to preprocess the detected text data.

[0153] The preprocessing module includes at least one or more of the following items 1)-4):

[0154] 1) A denoising submodule, used to denoise the detected text data;

[0155] 2) A completion submodule, used to complete sentences of the detected text data;

[0156] 3) A format unification submodule, used for format unification of the detected text data;

[0157] 4) Normalization submodule, used for normalizing the detected text data.

[0158] The reconstruction module 403 is used to reconstruct the detected text data in reverse order based on preset weights to obtain reconstructed text data.

[0159] Among them, in some embodiments of the present application, the reconstruction module includes:

[0160] An input submodule, used to input the preprocessed detection text data into a pre-trained reconstruction model; the reconstruction model includes a conversion module, a feature extraction module and a reverse reconstruction module;

[0161] A first acquisition submodule is used to acquire text vector data obtained by converting the detected text data by the conversion module;

[0162] The second acquisition submodule is used to obtain text feature information obtained by the feature extraction module through feature extraction of text vector data based on preset weights;

[0163] The third acquisition submodule is used to obtain the reconstructed text data obtained by the reverse reconstruction module through reverse reconstruction of the text feature information based on preset weights.

[0164] The calculation module 404 is used to calculate the maximum likelihood estimation value according to the detection text data and the reconstruction text data to obtain the abnormal probability of the power generation equipment.

[0165] The determination module 405 is used to determine the abnormality monitoring result of the power generation equipment according to the abnormality probability of the power generation equipment.

[0166] The determination module includes:

[0167] The fourth acquisition submodule is used to obtain a preset data abnormality probability threshold;

[0168] A first determination submodule is used to determine that the abnormality monitoring result of the power generation equipment is abnormal in response to the abnormality probability of the power generation equipment being greater than a preset data abnormality probability threshold;

[0169] The second determination submodule is configured to determine that the abnormality monitoring result of the power generation equipment is normal in response to the abnormality probability of the power generation equipment being less than or equal to a preset data abnormality probability threshold.

[0170] A pre-training module is used to pre-train the reconstruction model; the pre-training module is specifically used for:

[0171] Get the detection text sample data;

[0172] Preprocess the detection text sample data;

[0173] Input the preprocessed test text sample data into the target neural network model; the target neural network model includes a conversion module, a feature extraction module and a reverse reconstruction module;

[0174] Input the preprocessed detection text sample data into the conversion module to obtain text vector sample data;

[0175] Input text vector sample data into the feature extraction module to obtain text feature sample information based on preset weights;

[0176] Optionally, in some embodiments of the present application, the reconstruction model also includes a self-attention module; wherein the pre-training module is specifically used to: input text vector sample data into the self-attention module to obtain weight information of the text vector sample data; input the text vector sample data and weight information into the feature extraction module to obtain text feature sample information based on preset weights.

[0177] The text feature sample information is input into the reverse reconstruction module to obtain reconstructed text data that is reverse reconstructed based on preset weights.

[0178] Optionally, in some embodiments of the present application, the reconstruction model also includes a self-attention module; wherein the pre-training module is specifically used to: input text feature sample information into the self-attention module to obtain weight information of the text feature sample information; input the text feature sample information and weight information into the reverse reconstruction module to obtain reconstructed text data.

[0179] According to the abnormality monitoring device for power generation equipment of the embodiment of the present application, the detection text data is reversely reconstructed based on preset weights to obtain reconstructed text data, and the maximum likelihood estimate is calculated based on the detection text data and the reconstructed text data to obtain the abnormality probability of the power generation equipment. This can accurately predict the abnormality of the power generation equipment and improve the accuracy of abnormality monitoring of the power generation equipment.

[0180] According to an embodiment of the present application, the present application also provides an electronic device and a readable storage medium.

[0181] like Figure 5 , is a block diagram of an electronic device according to a method for monitoring abnormality of power generation equipment according to an embodiment of the present application. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.

[0182] like Figure 5 As shown, the electronic device includes: one or more processors 501, a memory 502, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are interconnected using different buses and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In other embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 5 A processor 501 is taken as an example.

[0183] The memory 502 is a non-transient computer-readable storage medium provided in the present application. The memory stores instructions executable by at least one processor to enable the at least one processor to perform the power generation equipment abnormality monitoring method provided in the present application. The non-transient computer-readable storage medium of the present application stores computer instructions, which are used to enable a computer to perform the power generation equipment abnormality monitoring method provided in the present application.

[0184] The memory 502 is a non-transient computer-readable storage medium that can be used to store non-transient software programs, non-transient computer executable programs and modules, such as program instructions / modules corresponding to the power generation equipment abnormality monitoring method in the embodiment of the present application (for example, the attached Figure 4 The processor 501 executes various functional applications and data processing of the server by running the non-transient software programs, instructions and modules stored in the memory 502, that is, the abnormality monitoring method of the power generation equipment in the above method embodiment is implemented.

[0185] The memory 502 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required by at least one function; the data storage area may store data created according to the use of the power generation equipment abnormality monitoring electronic device, etc. In addition, the memory 502 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory 502 may optionally include a memory remotely arranged relative to the processor 501, and these remote memories may be connected to the power generation equipment abnormality monitoring electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0186] The electronic device of the power generation equipment abnormality monitoring method may further include: an input device 503 and an output device 504. The processor 501, the memory 502, the input device 503 and the output device 504 may be connected via a bus or other means. Figure 5 The example of connecting through bus is taken in the following.

[0187] The input device 503 can receive input digital or character information, and generate key signal input related to user settings and function control of the power generation equipment abnormality monitoring electronic device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator rod, one or more mouse buttons, a trackball, a joystick and other input devices. The output device 504 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The display device may include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display and a plasma display. In some embodiments, the display device may be a touch screen.

[0188] Various implementations of the systems and techniques described herein can be realized in digital electronic circuit systems, integrated circuit systems, dedicated ASICs (application specific integrated circuits), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0189] These computer programs (also referred to as programs, software, software applications, or code) include machine instructions for programmable processors and can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, device, and / or means (e.g., disk, optical disk, memory, programmable logic device (PLD)) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

[0190] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0191] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0192] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short).

[0193] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this application can be executed in parallel, sequentially or in different orders, as long as the expected results of the technical solution disclosed in this application can be achieved, and this document is not limited here.

[0194] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A method for monitoring abnormality of power generation equipment, comprising: Get the detected text data; Preprocessing the detected text data; Reconstructing the detected text data in reverse order based on preset weights to obtain reconstructed text data; Calculating a maximum likelihood estimation value based on the detection text data and the reconstruction text data to obtain an abnormal probability of the power generation equipment; Determining an abnormality monitoring result of the power generation equipment according to the abnormality probability of the power generation equipment; The step of reconstructing the detected text data in reverse order based on preset weights to obtain reconstructed text data includes: Inputting the preprocessed detection text data into a pre-trained reconstruction model; the reconstruction model includes a conversion module, a feature extraction module and a reverse reconstruction module; Acquire text vector data obtained by converting the detected text data by a conversion module; Acquiring text feature information obtained by a feature extraction module performing feature extraction on the text vector data based on the preset weight; Obtaining reconstructed text data obtained by the reverse reconstruction module performing reverse reconstruction on the text feature information based on the preset weight; The reconstruction model is pre-trained by the following steps: Get the detection text sample data; Preprocessing the detection text sample data; Input the preprocessed detection text sample data into a target neural network model; the target neural network model includes a conversion module, a feature extraction module and a reverse reconstruction module; wherein the target neural network model is a long short-term memory network model; the conversion module is a Transformer conversion module; Inputting the preprocessed detection text sample data into the conversion module to obtain text vector sample data; Inputting the text vector sample data into the feature extraction module to obtain text feature sample information based on the preset weight; Inputting the text feature sample information into the reverse reconstruction module to obtain reconstructed text data that is reversely reconstructed based on the preset weights; The reconstruction model further includes a self-attention module; wherein the text vector sample data is input into the feature extraction module to obtain text feature sample information based on the preset weight, including: Inputting the text vector sample data into a self-attention module to obtain weight information of the text vector sample data; The text vector sample data and the weight information are input into the feature extraction module to obtain text feature sample information based on the preset weight.

2. The method according to claim 1, wherein: The step of inputting the text feature sample information into the reverse reconstruction module to obtain reconstructed text data that is reversely reconstructed based on the preset weights includes: Inputting the text feature sample information into a self-attention module to obtain weight information of the text feature sample information; The text feature sample information and the weight information are input into the reverse reconstruction module to obtain reconstructed text data based on the preset weights.

3. The method according to claim 1, wherein: Determining the abnormality monitoring result of the power generation equipment according to the abnormality probability of the power generation equipment includes: Obtain the preset data anomaly probability threshold; In response to the abnormal probability of the power generation equipment being greater than the preset data abnormal probability threshold, determining that the abnormal monitoring result of the power generation equipment is abnormal; In response to the abnormal probability of the power generation equipment being less than or equal to the preset data abnormal probability threshold, it is determined that the abnormal monitoring result of the power generation equipment is normal.

4. A power generation equipment abnormality monitoring device, comprising: An acquisition module is used to acquire detection text data; A preprocessing module, used for preprocessing the detected text data; A reconstruction module, used for reconstructing the detected text data in reverse order based on preset weights to obtain reconstructed text data; A calculation module, used for calculating a maximum likelihood estimation value according to the detection text data and the reconstructed text data to obtain an abnormal probability of the power generation equipment; A determination module, used to determine the abnormality monitoring result of the power generation equipment according to the abnormality probability of the power generation equipment; Wherein, the reconstruction module includes: An input submodule, used to input the preprocessed detection text data into a pre-trained reconstruction model; the reconstruction model includes a conversion module, a feature extraction module and a reverse reconstruction module; A first acquisition submodule is used to acquire text vector data obtained by converting the detected text data by the conversion module; The second acquisition submodule is used to obtain text feature information obtained by the feature extraction module through feature extraction of text vector data based on preset weights; The third acquisition submodule is used to obtain the reconstructed text data obtained by the reverse reconstruction module by reversely reconstructing the text feature information based on the preset weights; It also includes a pre-training module for pre-training the reconstruction model; the pre-training module is specifically used for: Get the detection text sample data; Preprocessing the detection text sample data; Input the preprocessed detection text sample data into a target neural network model; the target neural network model includes a conversion module, a feature extraction module and a reverse reconstruction module; wherein the target neural network model is a long short-term memory network model; the conversion module is a Transformer conversion module; Inputting the preprocessed detection text sample data into the conversion module to obtain text vector sample data; Inputting the text vector sample data into the feature extraction module to obtain text feature sample information based on the preset weight; Inputting the text feature sample information into the reverse reconstruction module to obtain reconstructed text data that is reversely reconstructed based on the preset weights; The reconstruction model further includes a self-attention module; the pre-training module is specifically used for: Inputting the text vector sample data into a self-attention module to obtain weight information of the text vector sample data; The text vector sample data and the weight information are input into the feature extraction module to obtain text feature sample information based on the preset weight.

5. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 3.

6. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 3.

7. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the steps of the method for monitoring abnormality of power generation equipment according to any one of claims 1 to 3.

Images