149results about How to "Realize intelligent identification" patented technology

The invention provides a network security method and system based on a block chain, and the method comprises the steps: collecting suspicious events reported by all block chain nodes through a block chain network on the basis of local anomaly detection, analyzing the data of a plurality of block chain nodes through a correlation analysis and time sequence mode mining algorithm, obtaining the correlation of different attack stages, and carrying out the detection of the local anomaly. Forming a composite attack graph under a global view angle; and according to key block chain nodes and paths in the composite attack graph, generating corresponding detection rules, issuing the detection rules to a local intelligence library of each block chain node, synchronously updating the rules of each block chain node by using the time sequence characteristics of the block chain, and triggering the smart contract to verify the validity of the newly added rules.

The invention provides a community management method and system based on an unmanned aerial vehicle, electronic equipment and a storage medium, and relates to the field of community management. The method comprises the steps of obtaining an initial inspection task of an unmanned aerial vehicle cluster and planning an inspection route, and controlling the unmanned aerial vehicle cluster to execute inspection operation to obtain multi-source inspection data. And constructing a community portrait based on the inspection data, identifying community management hidden dangers through a classification reasoning model, and generating an early warning event and a processing task. A processing task is decomposed into hierarchical sub-tasks by utilizing an ontology reasoning framework, a task execution main body is selected, and a task set is generated. And a multi-Agent reinforcement learning framework is adopted to carry out space-time collaborative arrangement on the task set to form an overall collaborative scheme, the overall collaborative scheme is converted into an interpretable scheduling decision instruction, and finally the unmanned aerial vehicle is controlled to execute the task. According to the technical scheme, the emergency response speed of security and protection and the capability of coping with complex conditions are improved.

The invention relates to the technical field of smart communities, and discloses a smart community target identification tracking method based on big data, which comprises the following steps: constructing an omnibearing monitoring network in a community, collecting real-time video and sensor data in the community, obtaining feature data of a target through an edge computing technology, and carrying out target identification tracking on the feature data; transmitting the target feature data to a cloud server through a network, identifying and classifying the target feature data, judging the type and identity of the target, analyzing the current behavior of the target according to a preset target historical behavior pattern library, and determining whether the behavior is abnormal or not; and if the behavior of the target is not matched with the normal mode in the historical behavior mode library, judging that the behavior is an abnormal behavior, and triggering an early warning mechanism. According to the invention, intelligent identification, dynamic tracking and timely early warning of the abnormal behavior of the smart community are realized, and the intelligent level and response efficiency of community safety management are improved.

The invention discloses an anti-tumor treatment adverse reaction risk early warning method and system, and relates to the technical field of medical information processing, and the method comprises the steps: obtaining the physiological index data of a patient, and obtaining a short-term risk prediction result based on the time sequence change characteristics of the physiological index data of a deep neural network; obtaining clinical feature data of the patient, constructing a susceptibility assessment model, and analyzing correlation features of the clinical feature data through the susceptibility assessment model to obtain a baseline risk assessment result; and according to the short-term risk prediction result and the baseline risk assessment result, calculating a comprehensive risk score, and when the comprehensive risk score exceeds an early warning threshold of the corresponding adverse reaction type, triggering an early warning signal. According to the method, a double-layer risk assessment system is constructed by combining instant physiological index changes and clinical features of patients, the technical problems that an existing early warning method is insufficient in consideration of individual differences of the patients and poor in early warning timeliness are solved, and the accuracy and timeliness of early warning of adverse reactions of anti-tumor treatment are improved.

The invention discloses an optical cable fault diagnosis system and method based on a vibration filtering algorithm, relates to the technical field of optical cable fault diagnosis, and solves the problems of low efficiency, long response time and the like in the prior art and the technical problem that the health condition of an optical cable is difficult to comprehensively assess through single optical power data. The quality data of the optical cable are acquired; training an artificial intelligence model based on the historical vibration signal data to obtain an optical cable fault identification model; recognizing the vibration data based on an optical cable fault recognition model to obtain a recognition result; calculating an optical cable aging evaluation coefficient based on the vibration signal data and the optical power loss data; constructing an optical cable aging evaluation coefficient correction model based on the environment data to correct the optical cable aging evaluation coefficient obtained through calculation, and obtaining a target optical cable aging evaluation coefficient; and judging whether to carry out fault early warning based on the identification result and the target optical cable aging evaluation coefficient. The technical problem is solved.

The invention discloses a power grid communication equipment fault identification method and system based on multi-source heterogeneous data characterization learning. The method comprises the following steps: firstly, collecting various complex multi-source heterogeneous data containing fault power grid communication equipment and preprocessing the complex multi-source heterogeneous data; extracting deep semantic features by using a pre-trained feature extraction network and obtaining a feature vector sequence; capturing time features of the feature vector sequence through a bidirectional long-short-term memory network, and obtaining complete sequence feature representation through a full connection layer; performing low-rank multi-modal fusion on the complete sequence feature representation in combination with cross-modal Transform to obtain three types of output splicing construction joint representation, and obtaining a fusion feature by using a self-attention mechanism; and finally, constructing a fault identification model by using the fusion features. Through innovative data preprocessing and data fusion technologies and in combination with an advanced deep learning model, fault-related characteristics can be automatically learned from multi-source heterogeneous data, and intelligent identification of the fault state of the power grid communication equipment is realized.

The invention relates to a visual analysis method and device for multiple internal corrosion failure types of a pipeline. The method comprises the following steps: inputting internal corrosion failure picture data to be recognized into an internal corrosion failure type recognition model to obtain a predicted internal corrosion failure type and a feature map of the last convolutional layer; analyzing the predicted internal corrosion failure type and the feature map of the last convolutional layer by using Grad-CAM to obtain a feature region thermodynamic diagram; based on the predicted internal corrosion failure type and the characteristic region thermodynamic diagram, obtaining the contribution degree of the characteristic of each characteristic region to an identification result; wherein the internal corrosion failure type identification model is obtained by the following steps: obtaining a pipeline internal corrosion failure data set; inputting the internal corrosion failure data set into an initial internal corrosion failure type identification model for training to obtain a trained internal corrosion failure type identification model; wherein the initial internal corrosion failure type identification model is constructed by using a Resnet18 convolutional neural network.

The embodiment of the invention belongs to the field of information technology and financial science and technology, and relates to a planting industry insurance evaluation method, which comprises the steps of obtaining satellite remote sensing data, an unmanned aerial vehicle image and ground sensor data of a target farmland, and integrating the satellite remote sensing data, the unmanned aerial vehicle image and the ground sensor data into comprehensive monitoring data of the target farmland, the comprehensive monitoring data comprises a multispectral image, multispectral data and topographic data; performing boundary identification according to the multispectral image and the topographic data to obtain boundary information of the target farmland; performing comprehensive analysis according to the multispectral image and the multispectral data, and determining crop type information and crop health state information of the target farmland; and evaluating the planting risk level of the target farmland according to the boundary information, the crop type information and the health state information, and generating a corresponding evaluation report. According to the invention, intelligent identification of farmland boundaries, crop types and crop health states is realized, and the accuracy and efficiency of planting industry insurance evaluation are improved.

The invention discloses an intelligent electricity demand identification system and method for a multi-element scene, and belongs to the field of power supply services, and the system comprises a scene presetting module which is used for presetting a plurality of electricity service scenes, and each electricity service scene is provided with at least one intention classification label; the model building module is used for building a natural language processing model; the intention recognition module is used for performing intention recognition on the customer expression text to obtain an intention classification result; the demand generation module is used for generating electricity demand information; the topology mapping module is used for establishing a topology mapping relationship between the WeChat group and the customer manager; and the information pushing module is used for automatically pushing the electricity demand information to a corresponding customer manager. The customer demand intention in the WeChat group is accurately recognized through the natural language processing model, and the technical effect of improving the power supply service efficiency and quality is achieved in combination with the topological mapping relation between the customer manager and the WeChat group.

The invention relates to the field of bridge detection, in particular to a bridge pavement layer damage detection method and system based on a ground detection radar. The method comprises the following steps: scanning a bridge pavement layer by using a ground detection radar, and collecting radar echo data; performing denoising and filtering processing to obtain a clear radar signal; judging whether a damage sign exists or not, if so, determining the radar signal as the damage sign, and entering the next step, otherwise, returning to the first step; waveform coding feature extraction is carried out on the detected damage signs, and damage features are identified; carrying out comparative analysis on the extracted damage features and a database in a normal state, and determining existence and types of the damage; according to an analysis result, utilizing an adaptive frequency modulation technology to accurately detect the damage depth and range; predicting a damage development trend; and obtaining a detection report according to the analysis result and the prediction result. The method improves the damage detection precision and detection efficiency, and achieves the accurate prediction of the damage development trend.

According to the hydraulic power plant electromechanical equipment state monitoring informatization management system provided by the embodiment of the invention, data monitoring and analysis are performed on pressure of a water inlet and a water outlet of a water turbine by adopting an artificial intelligence technology based on deep learning, and time sequence dynamic change modes of the pressure of the water inlet and the pressure of the water outlet in a local time domain are respectively extracted; and principal component feature joint coding is carried out on the water inlet pressure and the water outlet pressure in a local time domain to obtain pressure cooperative change information of the water inlet and the water outlet of the water turbine, and then global time domain aggregation is carried out on the pressure cooperative change information to realize intelligent identification of the leakage state of the water turbine. Therefore, the operation efficiency and the safety performance of the hydropower station can be effectively improved, and meanwhile the maintenance cost and the fault processing time are reduced.

The invention discloses a coal rock fracture image recognition method and system based on deep learning and post-processing, a coal rock fracture image recognition model is adopted to recognize and classify spliced coal rock fracture images, the model comprises an improved YOLOv5 target detection network, an image segmentation network and a classifier which are connected in sequence, after the image segmentation network outputs the binary image, parallax depth calculation is carried out on cracks and non-cracks in the crack images at different angles, and the cracks are determined based on the parallax depth; after the fracture is determined, carrying out image feature numeralization on the binarized image, and extracting fracture image feature parameters; the fractal dimension of the fracture is calculated, whether the fracture can be represented or not is preliminarily judged based on the fractal dimension, if yes, the fracture image feature parameters are input into a classifier, the type of the coal rock fracture is output, and if not, the specific type of the fracture is output. And the parallax depth, the fractal dimension and the image features are integrated for recognition and classification, so that the problems of misrecognition and missed recognition are effectively reduced.

The invention discloses a weld DR image defect intelligent identification method based on local-global efficient modeling improved YOLOv5. According to the method, an iRMBConv efficient feature extraction module is designed, and the module combines the technical advantages of dynamic global modeling and static local information fusion. In the Backbone network, the iRMBConv module can fully extract the multi-scale defect features of the weld DR image, and the recognition precision of the defects of different scales can be improved. Besides, in the Neck network, a context sensing module CSM is designed, the characteristics of different receptive fields are obtained by adopting expansion convolution with different expansion rates, the characteristic information of the tiny defects in the weld DR image is effectively enhanced, and then the recognition precision of the model on the tiny defects is improved. The intelligent identification model has high identification precision and good robustness, is suitable for the industrial fields of aviation, aerospace, nuclear industry, special equipment and the like, and promotes the application and deployment of the weld DR image defect intelligent identification technology.

The invention provides an ocean internal wave data acquisition method, medium and system, and belongs to the technical field of ocean hydrological observation.The ocean internal wave data acquisition method comprises the steps that an acoustic Doppler flow velocity profiler is used for acquiring water body movement data, a temperature-salinity-depth profiler is used for acquiring seawater parameters, and meanwhile a satellite altimeter and drifting buoy data are combined. Performing time-frequency analysis on the water body motion data, and obtaining internal wave candidate data by using band-pass filtering; and constructing an internal wave mathematical equation set comprising a hydromechanics basic equation, buoyancy frequency, a frequency dispersion relation and the like, and verifying and screening out effective data. And inputting the effective data into a pre-trained classification model to determine an internal wave type, finally extracting a key waveform segment through a preset extraction equation set, and generating and outputting internal wave characteristic waveform data. According to the method, multi-source observation data and a mathematical physical model are combined, identification, classification and feature extraction of the ocean internal waves are realized, and the problem that a data acquisition method capable of effectively extracting effective signals of the ocean internal waves and accurately identifying the features of the internal waves is lacked at present is solved.

The invention discloses a logging optical cable detection and evaluation method, system and device based on the Internet of Things, and relates to the technical field of artificial intelligence, and the method comprises the steps: collecting data in real time through the deployment of an Internet of Things sensor, and achieving the efficient transmission and preprocessing of data through the combination of a low-power wide area network technology; a deep learning model is utilized to accurately evaluate the health state of the optical cable, an unsupervised learning technology is utilized to automatically identify an abnormal signal, and finally, a maintenance strategy is intelligently recommended based on an abnormal signal detection result to realize optimal distribution of maintenance resources. The method has significant advantages in the aspects of abnormal signal identification and intelligent maintenance strategy recommendation, can significantly improve the accuracy and efficiency of optical cable detection and evaluation, reduces the maintenance cost, and has important practical value and wide application prospects.

The invention relates to the technical field of environment noise source type recognition, and particularly discloses an environment noise source type recognition method and device based on a deep convolutional neural network, and the method comprises the steps: collecting environment noise data through a microphone array, carrying out the signal processing of the environment noise data, and generating an environment noise Mel-spectrum image; and performing fine-grained spectral analysis on the Mel-spectrum image of the environmental noise data by using a deep convolutional neural network to capture deep semantic features of the environmental noise, thereby realizing intelligent recognition of the sound source type of the environmental noise data. Through the mode, the method can achieve the quick and accurate recognition of the types of the environmental noise sources, reduces the dependence on the design of artificial features, improves the recognition accuracy and efficiency, and provides powerful technical support for noise control and environmental protection.

The invention provides a pomacea canaliculata underwater intelligent identification and active capturing device which comprises a cabin, a capturing mechanical arm and a driving mechanism, the capturing mechanical arm is arranged at the front end of the advancing direction of the cabin, the tail end of the capturing mechanical arm is connected with the cabin, and the driving mechanism is arranged at the rear end of the advancing direction of the cabin. The universal cutting machine is arranged at the front end of the advancing direction of the cabin body, and a master control module, a battery module, a cabin inner container, a first lighting module, a first monitoring and recognition module, a sensor module, an opening and closing door and an exit are arranged in the cabin body; the cabin inner container is connected with an opening and closing door arranged at the top of the cabin body and an outlet arranged at the bottom of the cabin body through two cabin inner container channels. A second monitoring and recognition module and a second lighting module are further arranged at the front end of the advancing direction of the cabin body. According to the device, aquatic plants or sundries blocking the view field and the path are cleaned, the device freely moves underwater, automatic observation, intelligent recognition, capturing and storage are integrated, the labor cost is reduced, and the environment damage caused by manual capturing work is avoided.

The invention provides a laser ablation morphology online measurement and ablation feature recognition method, relates to the field of image processing, and solves the problems that an existing measurement method cannot adapt to tangential airflow interference, measurement distortion is caused by strong contrast of strong and weak light, and ablation features cannot be obtained online. The method comprises the following steps: a light supplementing measurement process: shooting a laser ablation morphology image by adopting a high-speed camera, and inhibiting strong self-illumination of an ablation area by adopting a mode of irradiating the ablation area by adopting a light supplementing light source, so as to carry out online measurement on the laser ablation morphology; in the image processing process, calibration parameters of the high-speed camera are obtained, preprocessing of a laser ablation morphology image is carried out, image correction is carried out in a perspective transformation mode, image segmentation is carried out in a background difference mode, an ablation area is detected, size measurement is carried out on the detected ablation area, and ablation feature recognition is completed; according to the invention, the online measurement of the ablation morphology and the intelligent identification of the ablation morphology characteristics can be effectively realized.

The invention relates to the technical field of medical informatization, and particularly discloses a blood disease intelligent diagnosis and treatment data processing system which adopts an image processing technology based on deep learning to perform cell detection on a bone marrow morphological image of a patient object so as to focus on a cell target in the image. The method comprises the following steps: acquiring a cell target, performing multi-level feature extraction on the cell target to capture superficial detail features and trunk features of the cell, and performing significant sparse interactive coding on the superficial detail features and the trunk features to obtain cell feature representation with higher distinction degree, thereby realizing intelligent identification of abnormal cells. Through the mode, the cell characteristics in the blood sample can be automatically identified and analyzed, and rapid detection and diagnosis of abnormal cells are realized, so that a doctor can be assisted to carry out faster and more accurate blood disease diagnosis, and the workload of the doctor is relieved.

The invention provides a data resource collecting and labeling method based on big data mining, and relates to the technical field of data collecting and labeling. The method comprises the specific implementation steps of collecting multi-view traffic images, traffic flow data and vehicle speed data of a target road section in real time, preprocessing the multi-view traffic images, and dividing the multi-view traffic images into traffic image groups according to a time sequence; secondly, calculating a traffic concentration index in the first view angle traffic image, identifying a blocking region, mapping the blocking region to another view angle traffic image, and performing congestion labeling; then, through projection of an overlapped detection frame and traffic element point cloud calculation, a traffic accident is marked in combination with a vehicle sag detection model; and finally, environment consistency verification and annotation conflict verification are carried out on a congestion annotation result to ensure the accuracy and integrity of the data. According to the method, the processing accuracy of the traffic data is improved, effective support is provided for traffic management and decision making, and the operation efficiency of an intelligent traffic system is optimized.

The invention relates to an overseas website access encrypted traffic detection method based on multi-scale information entropy modeling, and belongs to the technical field of communication. The method comprises the following steps: collecting and preprocessing network traffic, and extracting a data packet byte sequence, a length sequence and a time interval sequence; calculating a multi-scale information entropy matrix of the traffic sequence based on different time scales and packet window scales; inputting the information entropy matrix into a convolutional neural network and attention mechanism fusion method for feature extraction; a long-time dependency relationship is modeled through a self-attention mechanism by using a Transform model, so that the accuracy of flow analysis is further improved; model training optimization is carried out through a cross entropy loss function and an Adam optimization algorithm, and the generalization ability is improved by adopting regularization and data enhancement methods. According to the invention, intelligent identification of Chinese websites and non-Chinese websites is realized in an encrypted traffic environment.

The invention relates to the technical field of automobile part machining control, and discloses an automobile part production monitoring method and system, and the system comprises an automobile part machining welding spot feature recognition module, an automobile part machining welding spot state analysis module, and an automobile part spot welding process parameter adjustment module. According to actual coordinate parameters and theoretical coordinate parameters of the body-in-white part machining welding spots, coordinate error parameters of the body-in-white part machining welding spots and theoretical position adjusting coordinate parameters of the body-in-white part machining welding spots are accurately measured by combining numerical analysis; digital metering of spot welding machining position errors and spot welding machining theory adjusting coordinates of automobile body-in-white parts is achieved; and the automobile part generation monitoring end performs automobile part machining welding spot position parameter adjustment operation in a self-adaptive manner based on the body-in-white part machining welding spot theoretical position adjustment coordinate parameters in combination with the industrial robot control end and the spot welding machining equipment.

The invention discloses an RH one-key refining system based on machine vision, and belongs to the technical field of metallurgical engineering and intelligent manufacturing, the RH one-key refining system based on machine vision comprises a data acquisition module, a data processing module, a data storage module, a splashing prediction module and an automatic regulation and control module, the data acquisition module is used for acquiring images before and after splashing in a vacuum tank of the RH furnace and parameters of the RH furnace in an operation process, the data processing module is used for extracting texture features of the images of the vacuum tank of the RH furnace, and the data storage module is used for storing all acquired data. The splashing prediction module is used for recognizing and predicting the splashing phenomenon by building a splashing prediction model, the automatic regulation and control module is used for regulating the RH circulation flow, the oxygen supply flow and the vacuum degree, and real-time monitoring and prediction of splashing are achieved by building the splashing prediction model and replacing manual observation based on the actual working condition.

The invention relates to the technical field of big data processing, and particularly provides an intelligent method and system for identifying, classifying and pushing business opportunities of enterprises, and the method comprises the following steps: S1, carrying out the data collection and preprocessing; s2, carrying out business opportunity identification; s3, business opportunity classification; and S4, business opportunity pushing. Compared with the prior art, accurate business opportunity services can be provided according to the preferences and demands of the users, and the satisfaction degree of the users is improved.

The invention provides a geological disaster monitoring and early warning method based on AI vision, and relates to the technical field of geological disaster monitoring and early warning, and the method comprises the steps: obtaining geological disaster pictures, including daytime pictures and night pictures; marking the pictures to form a geological disaster picture data set; training an AI model through the picture data set to obtain a trained AI model; carrying out lightweight processing on the trained AI model, and putting the AI model into an edge computer; importing the shot pictures into an AI model in an edge computer to identify geological disaster features, identifying the geological disaster features by using a previous and later picture rolling comparison algorithm, and performing mutual verification on two results; and through mutual verification of the two results, the occurrence of the geological disaster or the geological disaster precursor is judged, and an early warning is given out. According to the invention, accurate monitoring and early warning of geological disasters are realized, and timeliness and effectiveness of disaster response are improved.

The invention relates to the technical field of geological disaster monitoring, in particular to a geological disaster monitoring method based on the 5G positioning technology, and the method comprises the steps: obtaining the geographic position information of a target region, and carrying out the displacement monitoring, underground space monitoring, underground water monitoring and meteorological monitoring of the target region based on the geographic position information, acquiring real-time displacement data, underground space detection data, underground water monitoring data and meteorological data; processing and analyzing the real-time displacement data, the underground space detection data, the underground water monitoring data and the meteorological data to obtain a geological disaster prediction result; and based on the geological disaster prediction result, giving an optimal treatment scheme and sending out early warning information. According to the invention, intelligent identification, early warning and treatment of hidden danger of geological disasters can be realized, and the safety of a monitoring area is effectively guaranteed.

The utility model discloses household garbage intelligent classification equipment with an automatic recognition function. The household garbage intelligent classification equipment comprises an aluminum profile frame, a two-degree-of-freedom cross shaft plane moving mechanism, a clamping mechanism, an opening and closing wing plate mechanism and an overturning type temporary storage table mechanism. Side plates are arranged on the side faces of the aluminum profile frame, and a feeding opening upper cover plate is arranged at the top. The aluminum profile frame is divided into two layers through a bottom-layer partition plate and a middle-layer aluminum profile beam, and a plurality of partition plates are arranged between the bottom-layer partition plate and the middle-layer aluminum profile beam to form a plurality of compartments for garbage classification; a compression mechanism is arranged between the bottom-layer partition plate and the middle-layer aluminum profile beam; the two-degree-of-freedom cross shaft plane mechanism is arranged below the throwing opening upper cover plate; the clamping mechanism is mounted in the two-degree-of-freedom cross shaft plane mechanism; the opening and closing wing plate mechanism is arranged below the two-degree-of-freedom cross shaft plane mechanism; the overturning type temporary storage table mechanism is installed on the middle-layer aluminum profile beam and located below the opening and closing wing plate mechanism.

The invention relates to the field of protector data supervision, in particular to an electric line protector operation fault supervision system and method based on big data. According to the invention, data preprocessing is carried out on the collected real-time operation parameter data of the power utilization line protector to generate standard real-time operation parameter data of the power utilization line protector, and numerical value matching is carried out on the standard real-time operation parameter data of the power utilization line protector and the operation parameter analysis data of the power utilization line protector generated according to the historical operation parameter data of the power utilization line protector; according to the method, whether the power utilization line protector breaks down or not is judged, operation fault processing scheme analysis is carried out on the power utilization line protector with the fault, target power utilization line protector fault solution data are generated, intelligent recognition of the power utilization line protector fault is achieved, and a fault solution is scientifically generated.

The present disclosure relates to a method and device for identifying tectonic coal based on borehole gamma data. The method comprises: acquiring gamma curve data and mechanical drilling speed data collected in real time during the process of logging while drilling; preprocessing the gamma curve data and the mechanical drilling speed data to obtain depth sequence data; performing segmentation processing on the depth sequence data to obtain a plurality of data sample units; for each data sample unit, extracting a plurality of sample features for representing the response mode of tectonic coal from the data sample unit, fusing the plurality of sample features to obtain a multi-parameter fusion feature vector; and inputting the multi-parameter fusion feature vector into a trained machine learning classification model to obtain a classification result output by the machine learning classification model. The present scheme improves the accuracy and efficiency of tectonic coal identification.

The invention provides a transformer winding looseness detection method and system, and the method comprises the steps: installing a plurality of acceleration sensors at each monitoring point position of a target transformer, so as to monitor an initial vibration signal of the target transformer in real time; de-noising the initial vibration signal collected by each monitoring point position by using a connection degree model, and fusing the de-noised information to obtain a pre-processing signal of the target transformer; performing normalization processing and time division on the preprocessed signal to obtain a sub-vibration time domain waveform signal; obtaining the maximum value of each sub-vibration time-domain waveform signal, and combining the maximum values to obtain a peak sequence signal; extracting characteristic parameters of the peak sequence signal by using a time-frequency analysis method, and inputting the characteristic parameters into the trained winding looseness identification model to obtain an identification result; and the structure of the winding looseness identification model is a CNN-LSTM network. The method realizes intelligent identification of winding looseness, can adapt to a complex vibration mode, and improves the identification efficiency and accuracy.