18586results about How to "Improve versatility" patented technology

The invention provides a twin power station collision visual simulation method and system based on physical simulation, and relates to the technical field of data processing, and the method comprises the steps: obtaining a three-dimensional geometric model of a twin power station, carrying out space division, extracting geometric parameters, obtaining material data, constructing a material correlation model and a mechanical model, and carrying out fusion to obtain a physical simulation model; acquiring initial collision condition parameters as boundary conditions, performing local grid encryption, constructing a nonlinear kinetic equation of a collision process, performing time domain solution, determining equivalent stress distribution and equivalent plastic strain distribution, and generating a collision result; judging whether plastic deformation or fracture failure occurs, generating full collision process physical field data, generating a collision animation frame sequence, generating a stress distribution cloud picture and a plastic strain cloud picture, constructing a fracture special effect, and comprehensively generating a collision visual scene.

The invention discloses an object labeling method and system based on laser point cloud, particularly relates to the technical field of laser point cloud processing, and is used for solving the problem of inaccurate labeling result caused by a complex reflection phenomenon caused by a non-Lambertian surface in the prior art. The method comprises the steps of obtaining original point cloud data collected by a laser radar and performing preprocessing; identifying a non-Lambertian reflection region based on the reflection intensity information and the geometric topology feature, and carrying out filtering correction; correcting false mirror image points through multi-layer iteration correction, and reconstructing boundaries to generate corrected point cloud data; grouping and classifying target objects in the corrected point cloud data based on a clustering algorithm combining global features and local features; finely dividing the boundary of the preliminary segmentation result by adopting an attention mechanism network, and evaluating the uncertainty of a boundary region through a local mutual information entropy; and adjusting a boundary division strategy according to an uncertainty evaluation result, and generating an optimized segmentation result.

The invention provides a model pre-training method and device for a multi-language task, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining a multi-modal training data set which comprises a plurality of pieces of sample text data and a plurality of pieces of sample image-text pair data with multi-language contents, and covers a universal field and a preset business field; based on a plurality of sample image-text pair data, performing comparative learning training of alignment of visual features and text features on the initial model, freezing model parameters of a decoding module and adjusting model parameters of a visual encoder and a projection module in the training process until a first ending condition is met; based on the multiple pieces of sample image-text pair data and the multiple pieces of sample text data, constraint training of content understanding is carried out on the initial model meeting a first ending condition, and model parameters of a visual encoder, a projection module and a decoding module are adjusted in the training process until a second ending condition is met to obtain a target model; the information extraction capability of the model in the specific field can be remarkably improved.

The invention relates to the technical field of industrial automation, and discloses a method for detecting abnormal operation of industrial control equipment based on industrial control protocol analysis, which comprises the following steps of: 1, acquiring operation data of the industrial control equipment through at least one industrial control protocol which is a Modbus protocol, an OPC UA protocol or a PROFIBUS protocol, the operation data comprises sensor data, register data and equipment state data of the equipment; and 2, analyzing the collected data packet by using a protocol analyzer, extracting key information related to the equipment operation state, the key information comprising parameters such as equipment working frequency, load, vibration, temperature and pressure, and converting the key information into a uniform data format. By collecting, analyzing and analyzing the operation data of the industrial control equipment in real time, abnormal conditions in operation of the equipment can be found in time, spreading of equipment faults is avoided, and stable operation of the equipment is guaranteed.

The invention discloses a denoising network model based on a diffusion model and an oracle rubbing image restoration method, which can improve the universality of the restoration method, improve the diversity of restoration results, reduce the calculation overhead and reduce the manual dependence. The denoising network model is adjusted based on an existing U-Net network and comprises a convolutional trunk network, an encoder, a middle layer, a decoder, a deep convolutional layer and a feedforward neural network which are connected in sequence; segmenting the input image by the convolutional trunk network; the encoder comprises a plurality of groups of cascaded paired visual state space blocks and image merging blocks; the decoder comprises a plurality of groups of cascaded paired visual state space blocks and image expansion blocks; performing down-sampling and up-sampling operations on the image merging block and the image expansion block respectively; the visual state space block adopts the architecture of an existing Mama module, and a two-dimensional selective scanning module is selected to replace an S6 module of the Mama module.

The invention discloses an air pollutant concentration prediction method and related equipment thereof, and relates to the technical field of data processing. The method comprises the following steps: acquiring current pollutant concentration data, current meteorological data and target site geographic position data corresponding to a target environment monitoring site; taking the current pollutant concentration data, the current meteorological data and the geographic position data of the target station as target input data of the prediction model; wherein an input module of the prediction model is used for performing modal decomposition processing on input data to obtain multiple pieces of modal data; the time feature extraction module is used for extracting time feature data; the spatial feature extraction module is used for extracting spatial feature data; the space-time fusion convolutional network module is used for extracting space-time features; the output module is used for outputting a prediction result; determining a prediction result of the air pollutant concentration corresponding to the target environment monitoring station according to the output result; the method achieves the effects of improving the comprehensiveness and accuracy of a prediction result and expanding application scenes.

The utility model relates to a multi-stroke wire pushing rod mechanism of a wire drawing machine, which belongs to the technical field of wire drawing machines and comprises a wire drawing machine frame, a plurality of machine heads are arranged at the front end of the wire drawing machine frame, yarns which are wound with the machine heads to form a barrel are arranged on the machine heads, and composite wire feeding rings which are fixedly connected with the machine heads are arranged at the front ends of the machine heads. A yarn pushing rod mechanism fixedly connected with a rack of the wire drawing machine through bolts is arranged above the machine head, and the yarn pushing rod mechanism drives yarn to complete the creeling and head feeding process on the composite yarn feeding ring. The device has the characteristics of compact structure, good operation stability and high use efficiency. The problem that the amount of yarn falling into a deep groove is large is solved. The problem that large and small silk threads cannot be automatically fed and automatically reeled on the same drawing machine is solved. The front end of the wire pushing air cylinder is jacked, and any position control is achieved through the stroke positions of the two air cylinders.

The invention relates to a tree neighbor context-based big language model RAG optimization method, which comprises the following steps of: reading a directory structure and content of a document, generating a corresponding directory tree, converting the directory tree into a Markdown format containing a title and a level, and automatically extracting according to a document template if no title exists; segmenting the document into knowledge blocks according to the priority; after the knowledge blocks are recalled, the context is expanded by using a tree adjacent context method, and a new knowledge block list is formed; grouping and sorting the expanded knowledge blocks according to document identifiers, splicing reference data based on different fields and confidence coefficients, and generating a character string set from the original document; and splicing prompt words by using the constructed reference data and the new knowledge block list, and sending the spliced prompt words to the large model for processing. According to the scheme, the performance and practicability of the large model in the information retrieval and answer generation process are comprehensively enhanced by optimizing context information, emphasizing key content, adapting to various document formats, improving user experience and the like.

The invention discloses an antenna polarization simulation method based on phase and amplitude calibration, and the method comprises the steps: collecting original data, carrying out the preprocessing, and obtaining the preprocessed original data; constructing an initial calibration model, and generating a calibrated lookup table; obtaining a target polarization parameter, and calculating a Stokes parameter; constructing a Mueller matrix, and determining an ideal polarization state; constructing a closed-loop control system to obtain a monitoring result; constructing a polarization parameter adjustment strategy, executing multi-objective optimization, and generating an optimized polarization parameter; acquiring position information of each antenna, and constructing a multi-antenna system model; a distributed optimization algorithm is used to obtain a coordination result, a space diversity strategy is realized, and polarization control parameters are generated; and generating a final polarization control parameter by using a beam forming technology. According to the invention, the universality of the system is improved, the precision and reliability of polarization simulation are improved, the adaptability and stability of the system are enhanced, the complexity of real-time calculation is reduced, and the response speed of the system is improved.

The invention provides a code peculiar smell detection method and system based on AST code measurement and code semantics. The method comprises the following steps of: firstly, constructing a FA-AST (Flow-Augmented Abstraction Syntax Tree) by analyzing a code snippet, and extracting three types of AST-based code metrics, namely a structure metric, a syntactic metric and a semantic metric from the FA-AST; then, a deep learning model is used for capturing contextual information and a dependency relationship of the code from the source code so as to extract code semantics; and finally, by fusing AST-based code measurement and code semantics, inputting a classification model to carry out training and testing so as to identify the peculiar smell in the code and evaluate the severity of the peculiar smell. Compared with software code measurement extracted by a traditional static code analysis tool, the AST-based code measurement provided by the invention is more comprehensive and easier to extract, and has obvious advantages in the aspect of improving the code odor detection precision and efficiency.

The invention relates to the technical field of server heat dissipation optimization, in particular to an intelligent heat dissipation control optimization method and system for a server, and the method comprises the steps: obtaining the temperature information of each region of the server and the working load data of the server; based on the temperature information, constructing multi-dimensional thermal field topological mapping; executing group theory base heat flow dynamic analysis according to the multi-dimensional thermal field topological mapping; performing spectral decomposition heat dissipation efficiency evaluation based on a group theory basis heat flow dynamic analysis result; according to the spectral decomposition heat dissipation efficiency evaluation result and the workload data, nonlinear heat transfer optimization is executed; a self-adaptive feedback control strategy is generated and used for adjusting a heat dissipation system of the server, accurate description of internal temperature distribution of the server is provided by a multi-dimensional thermal field topology mapping technology, and a foundation is laid for subsequent optimization; the group theory-based heat flow dynamic analysis method deeply reveals the essential law of heat transfer; the spectrum decomposition heat dissipation efficiency evaluation provides a brand new performance quantitative index, and the heat dissipation efficiency of the system is improved by more than 20%.

The invention relates to the technical field of bolt mounting equipment, in particular to an angle steel tower bolt perforation mounting method. By means of autonomous navigation of the installation robot and flexible operation of the mechanical arm, rapid positioning and taking-out of bolts and nuts are achieved, and the risk of manual climbing and operation is reduced. Accurate insertion and preliminary positioning of the bolt are ensured through accurate control of the mechanical arm, then the fastening effect is enhanced by combining impact tightening with the intermittent hammering technology, the fit clearance between the angle steel is effectively eliminated, and the bolt is prevented from loosening. The whole process is highly automatic, personal errors are reduced, and the mounting quality and the working safety are improved.

The invention relates to the technical field of intelligent lock recognition, in particular to an intelligent lock image feature recognition method and system based on deep learning. The recognition method comprises the following steps: constructing a multi-branch deep neural network model, wherein the multi-branch deep neural network model comprises a backbone network, a feature extraction branch and a feature fusion branch; a framework training model is used to learn identity recognition, attitude estimation and environment perception tasks; and identifying the comprehensive feature representation by using a trained framework training model, and generating a control instruction of the intelligent lock according to an identification result. According to the intelligent lock image feature recognition method based on deep learning provided by the invention, comprehensive analysis of multi-dimensional information such as identity, posture and environment is realized by constructing the multi-branch deep neural network model.

The invention relates to the technical field of integrated circuits, in particular to a processing system of a single-precision configurable matrix operation unit based on RISC-V instruction expansion, which comprises an RISC-V processor used for executing a software program and identifying a user-defined expansion instruction, and the RISC-V processor is connected with the single-precision configurable matrix operation unit through a user-defined coprocessor interface; the single-precision configurable matrix operation unit comprises an instruction response and decoding module and an instruction retirement and write-back module, and the instruction response and decoding module is used for receiving an instruction function code, source operation data and a destination register number and decoding the instruction function code; after decoding is completed, different execution units are selected according to a decoding result to send corresponding control signals; and after the execution unit completes the corresponding task, the instruction retirement and write-back module returns an execution result and a completion signal to the RISC-V processor, so that execution of a self-defined extension instruction is completed. According to the system, the matrix operation speed can be accelerated through bottom hardware under limited hardware resource consumption.

The invention relates to the technical field of motor shaft manufacturing, in particular to automatic milling equipment for motor shaft manufacturing. Comprising an equipment frame; a bottom plate is installed on the equipment frame, a milling mechanism used for milling a motor shaft from the side face is arranged on the rear side of the upper end of the bottom plate, and a fixing mechanism is arranged on the front side of the upper end of the bottom plate. According to the invention, the fixing mechanism firstly fixes and straightens the motor shaft, then the motor shaft is subjected to self-adaptive pressing through the pressing part, and then the milling mechanism mills the motor shaft from the side surface and cooperates with a cutting fluid, so that chippings generated in the milling process can be discharged from a key slot, and the damage to the key slot and a milling cutter is avoided; and the pressing part moves along with the milling mechanism in the milling process of the milling mechanism, so that the milling part of the milling cutter is relatively extruded and supported, the stability of the motor shaft is improved, vibration is reduced, and the milling quality of the key groove of the motor shaft is improved.

The invention belongs to the related technical field of construction safety inspection, and particularly relates to a construction site safety inspection method based on a multi-modal large language model, which comprises the following steps of: introducing the multi-modal large language model, and applying the multi-modal large language model for multiple times in combination with a knowledge base containing safety standard specifications and evaluation rules to obtain a construction site safety inspection result. The method specifically comprises the following steps: obtaining image content description through an image and a preset visual perception system prompt text; intelligently decomposing user input query into a plurality of sub-query tasks through user input query, image content description and preset task decomposition system prompt; searching construction safety management standard data related to each sub-query from the construction safety knowledge base, and analyzing and reasoning the sub-query based on the image content description in combination with the searched standard data to obtain the reasoning and conclusion of the sub-query; and finally, obtaining a total reasoning and conclusion based on each sub-query data. The method can accurately, scientifically and comprehensively realize the safety inspection of the construction site.

The invention discloses a feature classifier-based thinking chain reasoning mathematical problem solving method and system. The method comprises the following steps of: (1) acquiring a mathematical problem set, and selecting forward and reverse thinking chain examples to splice to form a new set; (2) respectively inputting the new set elements into a model for reasoning to construct a word-level reasoning path spanning tree, and selecting an attention weight matrix after pooling difference processing as node features for storage; (3) traversing all the generated reasoning path spanning trees, and screening nodes meeting requirements to construct a feature classifier training set; (4) training a feature classifier by using a support vector machine algorithm; and (5) through the trained feature classifier, participating in path selection in a pre-training language model reasoning process, and obtaining a relatively accurate reasoning process and answers. By means of the method, adjustment and control of finer granularity of the reasoning path of the pre-training language model can be achieved, accurate reasoning solving of the pre-training language model on mathematical problems is facilitated, and the generalization level of the pre-training language model is improved.

The invention provides a petrochemical pipeline corrosion defect detection device and method, and relates to the field of pipeline defect detection. The detection device comprises a mounting plate, and further comprises a battery box mounted on one side of the mounting plate, and the extension structure comprises a transmission assembly arranged on one side of the mounting plate, a rotating frame rotationally connected to the outer side of the transmission assembly and a push arm rotationally connected to the outer side of the transmission assembly and rotationally connected with the rotating frame; a connecting wire connected with the controller is mounted on one side of the electromagnetic ultrasonic thickness measuring probe; the servo motor drives the lead screw to rotate, the moving frame drives the rotating frame to turn over, meanwhile, the auxiliary wheels and the moving wheels are tightly attached to the interior of the pipeline, the size adjusting function is achieved, the device can easily adapt to pipelines with different inner diameters so as to deal with various complex working conditions, the universality and flexibility of the detection device are enhanced, and the detection efficiency is improved. The problem that the size of a pipeline corrosion detection device in the prior art is inconvenient to adjust is solved.

The invention belongs to the field of liquefied gas ships, and particularly relates to a gas test method of a liquefied carbon dioxide liquid cargo system for a ship. Comprising the following steps: drying a cargo hold space; liquid cargo tank and pipeline drying; pressurizing a liquid cargo pipeline; liquid cargo tank gas replacement; pressurizing the liquid cargo tank and the ventilation system; a liquid cargo tank cold tank; loading in a liquid cargo tank; testing the liquid cargo pump; unloading the liquid cargo tank; sweeping the liquid cargo tank; discharging the loading pipeline; heating the liquid collecting well; heating and ventilating the liquid cargo tank. According to the invention, the gas test process of the marine liquefied carbon dioxide liquid cargo system is simplified, and the operation efficiency is improved; before the liquid cargo tank gas replacement, pressurization and tank cooling are carried out, all the liquid phase pipelines are pressurized through dry gas, the gas test of the marine liquefied carbon dioxide liquid cargo system is carried out under the semi-cooling and semi-pressure condition, and the efficiency and safety of gas test operation are improved.

The invention discloses a manipulator for carrying automobile sheet metal parts, and relates to the technical field of manipulators, the manipulator comprises a supporting base, a mechanical arm and a plurality of suckers, the mechanical arm is mounted at the top of the supporting base, a U-shaped plate is mounted at the end, away from the supporting base, of the mechanical arm, and a supporting plate is fixedly connected to the lower end of the U-shaped plate; the supporting plate is provided with an adjusting assembly for automatically adjusting the height of the suction cups according to the automobile door frame, the adjusting assembly comprises limiting springs arranged on the outer surfaces of the suction cups in a sleeving mode, the middle of the U-shaped plate is hollow, and adjustment can be rapidly conducted according to the concave-convex condition of the door frame through the suction cups capable of being adjusted up and down; the time for manual intervention or readjustment due to the fact that the suction cups are not matched with the automobile door frame is shortened, the situation that adsorption is not stable due to the fact that the suction cups do not effectively make contact with the automobile door frame is avoided, and the phenomenon that abrasion occurs due to too large pressure when only a small number of suction cups are matched with clamping jaws to grab the automobile door frame is avoided; and the universality and the application range of the sucker type manipulator are improved.

The invention relates to the technical field of intelligent agriculture, and provides a negative pressure type fruit picking mechanical arm and a control system thereof.The negative pressure type fruit picking mechanical arm comprises a connecting chassis, a plurality of wheels are rotationally connected to the outer side wall of the connecting chassis, a storage groove is fixedly connected to the outer side wall of the top end of the connecting chassis, and a picking mechanism is arranged at the top end of the connecting chassis; the picking mechanism comprises a connecting frame which is mounted at the top end of the connecting chassis; by integrating an advanced mechanical arm and a visual recognition technology, the robot can accurately recognize and quickly pick crops, the picking efficiency is greatly improved, the robot is particularly suitable for large-scale agricultural production, the crop quality is improved, crop maturity judgment is conducted through deep learning models such as YOLOv5, immature crops are prevented from being picked by mistake, and the crop quality is improved. The quality of picked crops is guaranteed, the market value of agricultural products is improved, the labor cost is reduced, full-process automatic operation is achieved, the labor demand is reduced, the labor cost of agricultural production is reduced, and meanwhile the labor intensity of farmers is relieved.

The invention discloses a multi-modal emotion recognition method and system based on gating feature selection and expert modeling, and the method comprises the following steps: obtaining an original voice signal and an original text, and then respectively extracting voice features and text features of different levels; dynamically selecting the most suitable speech features and text features from the speech features and text features of different levels through a corresponding gating feature selection network, carrying out weighted fusion, obtaining the fused speech features and the fused text features, and splicing the fused speech features and the fused text features to obtain multi-modal fusion features; the method comprises the following steps: further capturing information of different modals in multi-modal fusion features through a modal hybrid expert Transform module, fusing the information again to obtain new multi-modal fusion features, and then processing the new multi-modal fusion features through an expert model which is most suitable for processing the new multi-modal fusion features in the hybrid expert Transform module. And obtaining a final prediction result and predicting the emotional state based on the prediction result. According to the method, the emotion prediction accuracy is improved.

The invention provides a network node classification method based on a multilayer graph neural network model, and the method comprises the steps: constructing a graph according to network nodes, carrying out the preprocessing of initial node features in the graph, and obtaining an initial embedded vector; according to the initial embedding vectors of the node and the neighbor node thereof, updating the node representation by adopting a layered message passing mechanism; according to the initial embedded vectors of the nodes and the neighbor nodes thereof, the graph attention network uses an attention mechanism to calculate weights for the neighbor nodes of each node and then dynamically updates feature vectors of the nodes; a multi-head attention mechanism is adopted to splice and merge the feature vectors of the nodes output by the graph attention network and the updated node representation, and the comprehensive feature vector of each node is obtained; the comprehensive feature vectors are further optimized through graph contrast learning and a node embedding module, and optimized features are obtained; and obtaining predicted node classification according to the optimization features. According to the method, complex features of heterogeneous graph data can be effectively captured, the method has good robustness for an over-smoothing problem, more accurate heterogeneous network representation learning is realized, and the method has advanced performance and robustness in a node classification task.

The invention relates to the technical field of hydrogen fuel cells, in particular to an air-cooled reactor automatic loading line for hydrogen fuel cell stack production, which comprises a rack, a storage rack, a material moving mechanism and two conveyor belts, the material moving mechanism comprises a support frame, a driving assembly and a pressing assembly, and electrode plates and bipolar plates are alternately conveyed to the material moving mechanism through the two conveyor belts. The electrode plate or the bipolar plate is supported through the sliding rollers at the bottoms of the overturning frames, and then the two overturning frames are synchronously driven to move through the driving assembly, so that the electrode plate or the bipolar plate naturally falls between the two overturning frames. The auxiliary electrode plate or the bipolar plate can stably fall on the storage rack in cooperation with downward pressing of the downward pressing assembly; the process does not need manual intervention, and the accuracy and regularity of stacking are ensured. The automatic stacking reduces the dependence on manual operation, reduces the labor cost, and further reduces the rework cost caused by unqualified materials due to the reduction of the surface damage of the electrode plates and the bipolar plates.

The invention discloses a fault root cause analysis method based on multi-source equipment data fusion. According to the method, multi-source heterogeneous data such as equipment operation data, maintenance records and environment data are collected and efficiently fused, a fault mode is automatically recognized from fused data, and time sequence evolution characteristics related to faults are mined. And based on the established knowledge base of the fault modes and reasons, and in combination with interactive Bayesian network reasoning and manual feedback, the root cause of the fault is deduced. According to the method, innovative technologies such as active learning and transfer learning are introduced, and the intelligent level, accuracy and efficiency of fault diagnosis are improved. And a visual interface visually presents an analysis process, supports man-machine interaction correction, and enhances interpretability. The method is suitable for the fields of industrial equipment, transportation tools, medical equipment and the like, and the system stability and usability are improved.

A detection method of a power transformer oil leakage detection system based on multi-modal prompt and multi-scale segmentation comprises the steps of data input and preprocessing, feature extraction, multi-modal feature fusion, Transformer decoder decoding, segmentation and classification, model training and fine adjustment and output and result display, and by combining Mask2Former and CLIP models, multi-modal information is fully utilized, and the detection result is more accurate. According to the method, the accuracy and robustness of oil leakage detection are improved, the characteristics of multi-modal information fusion, high-precision target segmentation, high robustness and high expandability are realized, accurate identification and segmentation of an oil leakage area are realized, other transformer fault detection tasks are adapted by simply modifying prompt words, and the method has good universality and adaptability.

The invention discloses a model wide-range attack angle change tail and web support device suitable for an airplane wind tunnel test, which comprises two support modes, namely a curved knife support rod and a web support support rod, an attack angle change mechanism, a lifting compensation mechanism and a mounting base, and is characterized in that a wind tunnel airplane model is arranged at the end part of the support rod, and the support rod is in screw connection with the attack angle change mechanism; the attack angle changing mechanism is connected to the lifting compensation mechanism through the mounting base, and the lifting compensation mechanism is mounted in wind tunnel test equipment. According to the invention, the large-range attack angle change of the airplane model in the wind tunnel test can be realized, and the center is kept unchanged at the height of the wind tunnel test.

The invention discloses an animal behavior recognition method and system based on deep learning. The method comprises the steps of data collection, data reconstruction, individual behavior classification, group correlation analysis and animal behavior recognition. The invention relates to the technical field of animal behavior recognition, in particular to an animal behavior recognition method and system based on deep learning, and adopts a comprehensive animal behavior recognition method combining individual analysis and group analysis to recognize a specific behavior type through individual behavior analysis. Comprehensive analysis and clustering are carried out by combining group correlation analysis with a plurality of animals, so that the recognition depth and comprehensive availability of animal behavior recognition are improved; performing individual behavior classification by adopting a residual convolutional recurrent neural network combined with structure and attention improvement; and performing group correlation analysis by adopting a dynamic and static graph neural integration network combined with reinforcement learning.

The invention provides an elevator door abnormal sound detection and identification method based on a variational auto-encoder (VAE) and a Transform network. The method comprises the steps of data acquisition and preprocessing, model construction and training, abnormal sound detection and identification, method performance evaluation and the like. The method specifically comprises the following steps: acquiring sound signals in the operation process of an elevator door by installing an audio sensor, continuously recording to form time sequence data, and cleaning and normalizing the time sequence data; then, feature reconstruction is carried out through a VAE-Transform network, a reconstruction error is calculated to obtain an abnormal score, abnormal feature vectors are continuously marked, finally, abnormal segments are input into a classifier, and the abnormal sound types are judged according to the confidence score of each abnormal sound type. According to the method, unsupervised network training is introduced to complete abnormal sound detection, and then the detected abnormal sound is further classified and identified, so that the calculation amount of data can be significantly reduced, the efficiency is greatly improved, the specific fault type is rapidly identified, and accurate maintenance suggestions are provided, thereby improving the fault processing efficiency and accuracy, and improving the fault processing efficiency and accuracy. The problem that abnormal sound fault data are scarce is effectively solved, and the method is suitable for abnormal sound fault monitoring of the elevator door in a complex environment.

The invention discloses a multi-mode-based self-adaptive control method and system for energy surgical equipment, and relates to the technical field of medical instrument control. The invention discloses a multi-mode-based self-adaptive control system for energy surgical equipment. The multi-mode-based self-adaptive control system comprises a waveform mode generation module, a real-time tissue parameter acquisition module, an acting tissue prediction module, a data analysis and decision module, a self-adaptive control output module and a safety monitoring and abnormal response module. By monitoring the contact area of the equipment electrode and the tissue in real time, the waveform mode and the output power are dynamically adjusted, and the uniformity and the high efficiency of energy transfer are ensured; on the basis of a tissue prediction model of individual characteristics of a patient, prior analysis is carried out by utilizing data acquired in real time, and response lag caused by drastic change of tissue states is avoided by predicting the tissue states at the next moment and adjusting waveform modes and parameters in advance, so that the prospective regulation and control capability of energy surgical equipment is remarkably improved.