Dam facility fault diagnosis method and device based on multi-modal cognition and medium

Through the acquisition and fusion of multimodal data, a deep learning model is built to diagnose dam facilities, solving the problems of incomplete information and low diagnostic accuracy caused by single modal data, and achieving high-precision and real-time fault diagnosis and early warning. Reduced maintenance costs.

CN118964972APending Publication Date: 2024-11-15山东浪潮智水数字科技有限公司 +1
View PDF 0 Cites 4 Cited by

Patent Information

Application Number
CN202410959556.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-17
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

Existing dam facilities fault diagnosis methods mostly rely on single modal data, resulting in incomplete information, low diagnostic accuracy, low diagnostic efficiency and high cost.

Method used

Using the fault diagnosis method of dam facilities based on multimodal cognition, the feature extraction and integration are performed by obtaining image data, sound data, vibration data, temperature data and displacement data, forming a target feature vector, and inputting the fault diagnosis model to output the fault Type, location and reason, and trigger the early warning mechanism.

Benefits of technology

It improves the accuracy and reliability of fault diagnosis, reduces manual intervention, improves diagnostic efficiency, reduces maintenance costs, and realizes intelligent fault diagnosis and real-time early warning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118964972A_ABST
    Figure CN118964972A_ABST
Patent Text Reader

Abstract

The invention discloses a dam facility fault diagnosis method and device based on multi-modal cognition and a medium, belongs to the technical field of intelligent water conservancy, and is used for solving the technical problems that an existing dam facility fault diagnosis method mostly depends on single-modal data, diagnosis information is not comprehensive and diagnosis precision is low. The method comprises the following steps: acquiring at least two different types of initial data of the dam facility; performing feature extraction on the initial data to obtain an initial feature vector corresponding to the initial data; integrating the initial feature vectors to form an integrated target feature vector; and inputting the target feature vector into a fault diagnosis model, and outputting the fault type of the dam facility, the position of the fault and the possible cause of the fault. According to the method, the accuracy and reliability of dam facility fault diagnosis are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of smart water conservancy technology, and in particular to a dam facility fault diagnosis method, equipment and medium based on multimodal cognition. Background Art

[0002] Dam facilities are an important part of water conservancy projects, and refer to dikes and weirs that intercept rivers and water, and dikes that intercept water in reservoirs and rivers. Dam facilities may experience different types of faults and faults in different locations during use. Therefore, it is necessary to diagnose faults in dam facilities, identify potential fault types, locations and causes in dam facilities, and predict fault development trends by analyzing and processing equipment operation data.

[0003] Existing dam facility fault diagnosis methods mostly rely on single modal data, such as vibration analysis or visual inspection, which have problems such as incomplete information, low diagnostic accuracy, and slow response speed. At the same time, traditional methods often require a lot of manual participation, low diagnostic efficiency and high cost. Summary of the invention

[0004] The present invention provides a dam facility fault diagnosis method, device and medium based on multimodal cognition, which are used to solve the technical problems that the existing dam facility fault diagnosis methods mostly rely on single modal data, the diagnostic information is incomplete and the diagnostic accuracy is low.

[0005] In a first aspect, the present invention provides a dam facility fault diagnosis method based on multimodal cognition, comprising:

[0006] Obtain at least two different types of initial data for the dam facility;

[0007] Performing feature extraction on the initial data to obtain an initial feature vector corresponding to the initial data;

[0008] Integrating the initial feature vectors to form an integrated target feature vector;

[0009] The target feature vector is input into a fault diagnosis model, and the fault type, the location of the fault and the possible cause of the fault of the dam facility are output.

[0010] In a feasible implementation manner, the at least two different types of initial data of the dam facility include: image data, sound data, vibration data, temperature data and displacement data.

[0011] In a feasible implementation manner, after acquiring at least two different types of initial data of the dam facility, the method further includes:

[0012] The initial data is preprocessed by denoising, data cleaning and data alignment.

[0013] In a feasible implementation manner, after acquiring at least two different types of initial data of the dam facility, the method further includes:

[0014] Performing image enhancement and edge detection on the image data;

[0015] Performing denoising and spectrum analysis on the sound data;

[0016] For the vibration data, temperature data and displacement data, outlier removal and feature engineering processing are performed.

[0017] In a feasible implementation manner, performing feature extraction on the initial data to obtain an initial feature vector corresponding to the initial data includes:

[0018] A deep learning large model is constructed, and features of the initial data are extracted through a convolutional neural network and / or a recurrent neural network to obtain an initial feature vector corresponding to the initial data.

[0019] In a feasible implementation, the initial feature vectors are integrated to form an integrated target feature vector, including:

[0020] The initial feature vectors are integrated using an attention mechanism or a multimodal fusion technique to form an integrated target feature vector.

[0021] In a feasible implementation manner, before inputting the target feature vector into the fault diagnosis model to output the fault type, the location of the fault and the possible cause of the fault of the dam facility, the method further includes:

[0022] Based on the target feature vector, a fault diagnosis model is constructed using a support vector machine;

[0023] The fault diagnosis model is trained by using labeled fault sample data to optimize the parameters of the fault diagnosis model.

[0024] In a feasible implementation manner, after inputting the target feature vector into the fault diagnosis model and outputting the fault type, the location of the fault and the possible cause of the fault of the dam facility, the method further includes:

[0025] The early warning mechanism is triggered to output early warning information to the user based on the fault type, the location of the fault and the possible cause of the fault.

[0026] In a second aspect, the present invention provides a dam facility fault diagnosis device based on multimodal cognition, the device comprising: at least one processor; and 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 so that the at least one processor can execute the dam facility fault diagnosis method based on multimodal cognition described in any of the above-mentioned embodiments.

[0027] In a third aspect, the present invention provides a non-volatile computer storage medium, which is a non-volatile computer-readable storage medium, and the non-volatile computer-readable storage medium stores at least one program, each of which includes instructions, and when the instructions are executed by a terminal, the terminal executes the dam facility fault diagnosis method based on multimodal cognition described in any of the above-mentioned embodiments.

[0028] The dam facility fault diagnosis method, device and medium based on multimodal cognition provided by the present invention have the following beneficial technical effects compared with the prior art:

[0029] (1) In the present invention, various types of initial data are used to obtain corresponding initial feature vectors through feature extraction. The target feature vector is formed by these initial feature vectors, which can more comprehensively reflect the operating status of the dam facilities and improve the accuracy and reliability of fault diagnosis.

[0030] (2) The present invention performs initial data acquisition, feature extraction, and fault diagnosis through a fault diagnosis model, thereby realizing intelligent fault diagnosis of dam facilities, reducing manual intervention, and improving the efficiency of fault diagnosis of dam facilities.

[0031] (3) After determining the fault type, location and possible cause, the present invention triggers an early warning mechanism and outputs early warning information to the user, providing managers and other users with sufficient time to take measures to prevent the expansion of faults in dam facilities.

[0032] (4) Through the intelligent fault diagnosis and real-time early warning of faults of the present invention, the maintenance cost of dam facilities is reduced and the economic benefits of dam facilities are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0034] Figure 1 A flow chart of a dam facility fault diagnosis method based on multimodal cognition provided by the present invention;

[0035] Figure 2A schematic diagram of the structure of a dam facility fault diagnosis device based on multimodal cognition provided by the present invention. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0037] Before the present invention is described in detail, terms that are or may be used in the present invention are explained first.

[0038] Large model: refers to a deep learning model with large-scale parameters and highly complex structure, which can process and analyze multi-source heterogeneous data and has strong learning and reasoning capabilities.

[0039] Multimodal cognitive capability: refers to the model’s ability to simultaneously understand and process information from different modalities (such as images, sounds, sensor data, etc.), and to perform effective cross-modal fusion to form a comprehensive understanding of the target object.

[0040] Fault diagnosis: The process of identifying potential fault types, locations and causes in dam facilities and predicting fault development trends by analyzing and processing equipment operation data.

[0041] The present invention realizes intelligent, high-precision and real-time diagnosis of dam facility failures by introducing the multimodal cognitive ability of a large model. The technical solution proposed by the present invention is described in detail below with reference to the accompanying drawings.

[0042] Figure 1 The flowchart of the dam facility fault diagnosis method based on multimodal cognition provided by the present invention. Figure 1 As shown, the method includes the following execution steps:

[0043] S101. Obtain at least two different types of initial data of the dam facility.

[0044] The present invention collects different types of initial data for dam facilities. These different types of initial data are multimodal data, which can reflect the operating status of the dam facilities in different dimensions and in different ways.

[0045] As a possible implementation manner, at least two different types of initial data of the dam facility include: image data, sound data, vibration data, temperature data and displacement data.

[0046] The present invention can take pictures of the dam facilities through a camera to collect image data of the dam facilities; collect sound data generated by the dam facilities through a sound sensor, collect vibration data generated by the dam facilities through a vibration sensor, collect temperature data generated by the dam facilities through a temperature sensor, and collect displacement data generated by the dam facilities through a displacement sensor.

[0047] As a possible implementation manner, after acquiring at least two different types of initial data of the dam facility, the method further includes: performing preprocessing including denoising, data cleaning and data alignment on the initial data.

[0048] After collecting multimodal initial data, the present invention can first pre-process the multimodal initial data, filter out noise on the initial data, and obtain denoised initial data; check the consistency of the initial data, process invalid values ​​and missing values ​​in the initial data, complete data cleaning of the initial data, and convert dirty data into data that meets data quality requirements; perform data alignment processing on the initial data to obtain initial data that meets the data alignment rules of the processor.

[0049] After the initial data is preprocessed by denoising, data cleaning and data alignment, the preprocessed initial data is obtained.

[0050] As a possible implementation method, after obtaining at least two different types of initial data of the dam facility, the present invention performs different preprocessing on the different types of initial data: for image data, image enhancement and edge detection are performed to obtain image data after image enhancement and edge images corresponding to the image data; for sound data, denoising and spectrum analysis are performed to obtain denoised sound data and spectrum analysis results of the sound data; for sensor data such as vibration data, temperature data and displacement data, outlier removal and feature engineering processing are performed.

[0051] S102: Extract features from the initial data to obtain an initial feature vector corresponding to the initial data.

[0052] The present invention converts the initial data into an initial feature vector suitable for processing by a machine learning algorithm through feature extraction, retains the key information in the initial data, reduces the data dimension, and ensures the quality and availability of the initial feature vector.

[0053] As a possible implementation method, a large deep learning model is constructed to extract features from the initial data through structures such as Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) to obtain the initial feature vector corresponding to the initial data.

[0054] After preprocessing the multimodal initial data, the present invention performs feature extraction on the preprocessed multimodal initial data to obtain initial feature vectors corresponding to each initial data.

[0055] Exemplarily, feature extraction is performed on the preprocessed image data to obtain an image feature vector; feature extraction is performed on the preprocessed sound data to obtain a sound feature vector; feature extraction is performed on the preprocessed vibration data to obtain a vibration feature vector; feature extraction is performed on the preprocessed temperature data to obtain a temperature feature vector; feature extraction is performed on the preprocessed displacement data to obtain a displacement feature vector.

[0056] S103, integrating the initial feature vectors to form an integrated target feature vector.

[0057] After extracting initial feature vectors corresponding to the multimodal initial data based on the multimodal initial data, the present invention fuses the initial feature vectors corresponding to the multimodal initial data to form a target feature vector, so as to achieve the complementarity of heterogeneous information in the multimodal initial data.

[0058] The target feature vector formed by multiple initial feature vectors contains rich information and can represent the feature information in the multimodal initial data, including the feature information of different types of initial data and the fused feature information obtained by fusion of multiple types of initial data.

[0059] As a possible implementation method, the initial feature vectors are integrated using attention mechanism or multimodal fusion technology to form a target feature vector.

[0060] As described in the above example, after performing feature extraction on the preprocessed image data to obtain an image feature vector; performing feature extraction on the preprocessed sound data to obtain a sound feature vector; performing feature extraction on the preprocessed vibration data to obtain a vibration feature vector; performing feature extraction on the preprocessed temperature data to obtain a temperature feature vector; and performing feature extraction on the preprocessed displacement data to obtain a displacement feature vector, the initial feature vectors such as the image feature vector, the sound feature vector, the vibration feature vector, the temperature feature vector, and the displacement feature vector can be fused to obtain the target feature vector.

[0061] In this way, the target feature vector can represent the feature information in the image data, sound data, vibration data, temperature data and displacement data, as well as the fused feature information obtained by fusing these initial data; compared with the feature vector obtained from the data of a single modality, the feature information represented by the target feature vector is more comprehensive, and can represent the fused feature information obtained by fusing multiple types of initial data.

[0062] S104, input the target feature vector into the fault diagnosis model, and output the fault type of the dam facility, the location of the fault, and the possible cause of the fault.

[0063] After obtaining the target feature vector corresponding to the dam facility, the present invention inputs the target feature vector into a fault diagnosis model, and the fault diagnosis model identifies the target feature vector to determine the fault type of the dam facility, the location of the fault and the possible cause of the fault.

[0064] The types of failures of dam facilities include: dam collapse failures due to excessively high water levels and unstable foundations, which lead to partial or complete collapse of the dam; cracks and deformations that occur during the construction and operation of dam facilities; leakage and leak failures caused by inadequate anti-seepage measures and poor construction quality; earth-rock dams are prone to landslides when floods overtopping, wind and waves hit, etc.; problems with the dam's drainage system or anti-seepage layer cause water to penetrate into the dam, resulting in infiltration and pipe bursts; aging and disease problems such as concrete spalling and steel corrosion due to long-term use and improper maintenance; culvert failures due to improper design, construction and material selection, as well as lack of regular inspections, which lead to the gradual expansion of small seepage and leakage problems.

[0065] The target feature vector is formed by the initial feature vectors corresponding to the various types of initial data. The target feature vector contains the feature information of the various types of initial data. That is to say, the feature information of the dam facility represented by the target feature vector is more comprehensive. By using the target feature vector that represents more comprehensive feature information to perform fault diagnosis on the dam facility, the accuracy and reliability of fault diagnosis can be improved.

[0066] As a possible implementation method, before inputting the target feature vector into the fault diagnosis model and outputting the fault type, location and possible cause of the dam facility, the method also includes: based on the target feature vector, using deep learning classifiers such as support vector machines (SVM) to build a fault diagnosis model; annotating the initial data with the fault type, fault location, and possible cause of the fault to obtain annotated fault sample data; training the fault diagnosis model with the annotated fault sample data, optimizing the parameters of the fault diagnosis model, and improving the accuracy of the fault diagnosis model in diagnosing the dam facility.

[0067] When a dam facility fails, it will affect the safe operation of the dam facility; and, if the failure of the dam facility is not repaired in time, it may cause the failure of the dam facility to expand, leading to more serious consequences; therefore, timely discovery and identification of the failure of the dam facility is particularly important for the safe management of the dam facility.

[0068] As a possible implementation method, in order to promptly eliminate the safety hazards caused by the failure of dam facilities, after inputting the target feature vector into the fault diagnosis model and outputting the fault type, location and possible cause of the dam facility failure, the method also includes: triggering an early warning mechanism based on the fault type, location and possible cause, and outputting early warning information to the user, so that the user can determine the fault type, location and possible cause in the fault diagnosis model. The user can be informed of the fault of the dam facility in a timely manner and take targeted measures according to the fault type, location and possible cause.

[0069] Exemplarily, fault diagnosis of dam facilities is achieved through a data acquisition module, a data preprocessing module, a cross-modal feature integration module, a fault model building module, and a fault diagnosis and early warning module.

[0070] Data acquisition module: Design and deploy a multimodal data acquisition system, including high-definition cameras, sound sensors, vibration sensors, temperature sensors, displacement sensors, etc., to collect multimodal operating data such as images, sounds, vibrations, temperature and displacement of dam facilities in real time, and transmit them to the data processing center.

[0071] Data preprocessing module: preprocesses the collected multimodal data, including denoising, data cleaning, feature extraction, and data alignment. For image data, image enhancement and edge detection are performed; for sound data, denoising and spectrum analysis are performed; for sensor data, outlier removal and feature engineering are performed.

[0072] Cross-modal feature integration module: Build a deep learning model to extract features from pre-processed multimodal data through structures such as convolutional neural networks and recurrent neural networks. Use attention mechanisms or multimodal fusion technology to effectively integrate feature vectors of different modalities to form a target feature vector containing rich information.

[0073] Fault diagnosis model building module: Based on the target feature vector, the fault diagnosis model is built using support vector machine, deep learning classifier and other algorithms. The model is trained using labeled fault sample data to optimize model parameters and improve the accuracy of fault diagnosis.

[0074] Fault diagnosis and early warning module: collects multimodal operation data of dam facilities in real time, processes it through the cross-modal feature integration module and inputs it into the trained fault diagnosis model. The model outputs the fault type, location and possible cause based on the input data, and triggers the early warning mechanism to notify the management personnel to take timely measures.

[0075] The present invention integrates multiple technologies such as image recognition, sound analysis, and sensor data processing. By constructing a large deep learning model, it realizes real-time collection, feature extraction, cross-modal fusion and intelligent diagnosis of multimodal operation data of dam facilities, significantly improving the accuracy and timeliness of fault diagnosis, and providing a strong guarantee for the safe and stable operation of dam facilities.

[0076] The present invention can more comprehensively reflect the operating status of dam facilities through the collection and fusion of multi-modal data, improve the accuracy and reliability of fault diagnosis, realize intelligent diagnosis of dam facility faults, reduce manual intervention, and improve diagnostic efficiency.

[0077] Moreover, once signs of failure are detected, the system can immediately trigger an early warning mechanism, providing managers with sufficient time to take measures to prevent the failure from expanding.

[0078] In this way, maintenance costs can be reduced and the economic benefits of dam facilities can be improved through intelligent diagnosis and real-time early warning.

[0079] The present invention includes the following key technical points:

[0080] (1) Implementation of multimodal data acquisition and processing technology to ensure the comprehensiveness and accuracy of data.

[0081] (2) Construction and training of large deep learning models, especially optimization of cross-modal feature extraction and fusion algorithms.

[0082] (3) Establishment and verification of fault diagnosis model to ensure the high accuracy and robustness of the model.

[0083] (4) Design and implementation of real-time warning mechanism to ensure timely transmission and processing of fault information.

[0084] Corresponding to the above-mentioned embodiments, the present invention also provides a dam facility fault diagnosis device based on multimodal cognition. Figure 2 A structural schematic diagram of a dam facility fault diagnosis device based on multimodal cognition provided in an embodiment of the present invention, the device may include: at least one processor; and 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 so that the at least one processor can execute the dam facility fault diagnosis method based on multimodal cognition as described in the above embodiment.

[0085] In a possible implementation of the present invention, the at least one processor is capable of executing the following steps: obtaining at least two different types of initial data of the dam facility; performing feature extraction on the initial data to obtain an initial feature vector corresponding to the initial data; integrating the initial feature vectors to form an integrated target feature vector; and inputting the target feature vector into a fault diagnosis model to output the fault type of the dam facility, the location of the fault, and the possible causes of the fault.

[0086] In a specific implementation, the present invention can also be provided as a non-volatile computer storage medium, wherein the storage medium is a non-volatile computer-readable storage medium, and the non-volatile computer-readable storage medium stores at least one program, each program including instructions, which, when executed by a terminal, enable the terminal to execute the dam facility fault diagnosis method based on multimodal cognition as described in the above embodiment.

[0087] In a possible implementation of the present invention, the aforementioned terminal executes to obtain at least two different types of initial data of the dam facility; performs feature extraction on the initial data to obtain an initial feature vector corresponding to the initial data; integrates the initial feature vectors to form an integrated target feature vector; inputs the target feature vector into a fault diagnosis model, and outputs the fault type of the dam facility, the location of the fault, and the possible causes of the fault.

[0088] The various embodiments of the present invention are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0089] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0090] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A dam facility fault diagnosis method based on multimodal cognition, characterized in that: The method comprises: Obtain at least two different types of initial data for the dam facility; Performing feature extraction on the initial data to obtain an initial feature vector corresponding to the initial data; Integrating the initial feature vectors to form an integrated target feature vector; The target feature vector is input into a fault diagnosis model, and the fault type, the location of the fault and the possible cause of the fault of the dam facility are output.

2. The dam facility fault diagnosis method based on multimodal cognition according to claim 1 is characterized in that: The at least two different types of initial data of the dam facility include: image data, sound data, vibration data, temperature data and displacement data.

3. The dam facility fault diagnosis method based on multimodal cognition according to claim 1 is characterized in that: After acquiring at least two different types of initial data of the dam facility, the method further comprises: The initial data is preprocessed by denoising, data cleaning and data alignment.

4. The dam facility fault diagnosis method based on multimodal cognition according to claim 2 is characterized in that: After acquiring at least two different types of initial data of the dam facility, the method further comprises: Performing image enhancement and edge detection on the image data; Performing denoising and spectrum analysis on the sound data; For the vibration data, temperature data and displacement data, outlier removal and feature engineering processing are performed.

5. The dam facility fault diagnosis method based on multimodal cognition according to claim 1 is characterized in that: Performing feature extraction on the initial data to obtain an initial feature vector corresponding to the initial data includes: A deep learning large model is constructed, and features of the initial data are extracted through a convolutional neural network and / or a recurrent neural network to obtain an initial feature vector corresponding to the initial data.

6. The dam facility fault diagnosis method based on multimodal cognition according to claim 1 is characterized in that: Integrating the initial feature vectors to form an integrated target feature vector includes: The initial feature vectors are integrated using an attention mechanism or a multimodal fusion technique to form an integrated target feature vector.

7. The dam facility fault diagnosis method based on multimodal cognition according to claim 1 is characterized in that: Before inputting the target feature vector into the fault diagnosis model to output the fault type, the location of the fault and the possible cause of the fault of the dam facility, the method further includes: Based on the target feature vector, a fault diagnosis model is constructed using a support vector machine; The fault diagnosis model is trained by using labeled fault sample data to optimize the parameters of the fault diagnosis model.

8. The dam facility fault diagnosis method based on multimodal cognition according to claim 1 is characterized in that: After inputting the target feature vector into the fault diagnosis model and outputting the fault type, the location of the fault and the possible cause of the fault of the dam facility, the method further includes: The early warning mechanism is triggered to output early warning information to the user based on the fault type, the location of the fault and the possible cause of the fault.

9. Dam facility fault diagnosis equipment based on multimodal cognition, characterized in that: The device comprises: at least one processor; and, 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, so that the at least one processor can execute the dam facility fault diagnosis method based on multimodal cognition according to any one of claims 1-8.

10. A non-volatile computer storage medium, characterized in that: The storage medium is a non-volatile computer-readable storage medium, which stores at least one program, each of which includes instructions, and when the instructions are executed by the terminal, the terminal executes the dam facility fault diagnosis method based on multimodal cognition according to any one of claims 1-8.

Citation Information

Cited By

  • Multi-modal large model construction and operation method and system for reservoir dam safety

    CN119622557A

  • A method and system for constructing and operating a multi-modal large model for reservoir dam safety

    CN119622557B

  • Data preprocessing method and device based on AI large model

    CN119645989A

  • Accurate dynamic risk assessment method for dike piping danger

    CN120067579A