339 results about "Online identification" patented technology

The invention discloses a dynamic monitoring-based deaerator water supply control method and system for a nuclear power plant, and relates to the field of deaerator water supply control, and the method comprises the steps: firstly, through a hybrid online identification algorithm, analyzing a historical valve position instruction and a water supply flow sequence in real time, and dynamically estimating the local gain and hysteresis width of a valve under the current working condition; then, the parameters identified in real time are utilized to carry out prospective compensation on the expected flow variation calculated by the main controller. In other words, the instruction amplitude is adjusted according to the estimated local gain, and it is ensured that consistent flow response can be obtained under different loads; meanwhile, when the instruction is reverse, the compensation amount is actively applied according to the estimated hysteresis width so as to eliminate the adjustment dead zone and oscillation caused by hysteresis. In this way, the nonlinear object of the valve is equivalent to a linear link with consistent response, and the stability and accuracy of the deaerator water level control system in the full working condition range are remarkably improved.

The invention provides a motor parameter online identification method based on an improved golden section method, and belongs to the technical field of motor parameter identification, and the method comprises the steps: S1, obtaining a residual function of a recursive least square method with a forgetting factor when parameter identification is not carried out on a motor model for the first time; s2, detecting whether the residual function exceeds a preset threshold value or not, and executing a step S3 when the residual function exceeds the preset threshold value; s3, taking the residual function as a target function, and adjusting the forgetting factor by using a golden section method to obtain an optimized forgetting factor; and S4, carrying out motor parameter online identification by using a recursive least square method with the optimized forgetting factor. The method has the advantages that the forgetting factor is dynamically adjusted through the golden section method, the residual function serves as the target function, the optimal solution of the forgetting factor under the corresponding working condition can be found when the system changes rapidly, the tracking speed is increased, the steady-state precision is further improved, and therefore the overall performance of parameter identification is improved.

The invention relates to the technical field of underground engineering safety monitoring and disaster simulation, in particular to a digital twinborn deduction platform for underground engineering disaster chain evolution simulation, which comprises a multi-disaster coupling numerical simulation module used for acquiring multi-source data including design drawing data, geological data and engineering monitoring data, and based on the multi-source data, establishing an initial stress field model by using a numerical calculation method combining a finite element and a finite difference, and carrying out leakage-settlement-structure failure multi-disaster coupling numerical simulation. According to the method, the partial differential equation for describing the evolution law of the underground engineering is used as a constraint term to be embedded into the deduction model, and online identification and dynamic extrapolation of parameters are executed by fusing real-time monitoring data, so that the accuracy of disaster evolution simulation under complex working conditions is remarkably improved; and the problem that the traditional numerical calculation method is difficult to meet the real-time requirement of digital twinning is solved.

The invention relates to a low-voltage transformer area phase sequence recognition graph connection prediction method fusing time sequence coding and graph attention. The method comprises the following steps: step S101, acquiring voltage time sequence data of all users; step S102, converting a traditional'node classification 'problem of allocating a phase label to each user into a'graph connection prediction' problem of judging whether any two users are connected to the same phase; step S103, constructing an end-to-end deep learning model; step S104, constructing a pairwise prediction module (MLP classifier); and step S105, deploying the finally trained model to an embedded computing device, and performing low-cost and high-efficiency online identification and effect test on the user phase relationship of an unknown transformer area in an actual or simulated environment. The method gets rid of dependence on physical parameters of a power grid, and realizes high-precision automatic phase identification through an innovative problem modeling mode and a deep learning architecture.

The invention belongs to the technical field of industrial control, and particularly relates to a brushless direct current motor high-efficiency control method and system based on an FOC algorithm, and the method comprises the steps: superposing high-frequency sine disturbance in a voltage instruction, extracting a response component, achieving the online identification of the equivalent impedance of a motor, and dynamically solving and updating the estimated values of resistance and inductance; and correcting d-axis and q-axis voltage feed-forward items in real time by using the updated parameters, and correspondingly adjusting a current loop proportion and an integral gain. The method can adapt to the time-varying characteristics of motor parameters in an industrial field, so that the accuracy of feed-forward compensation and the stability of current loop bandwidth are effectively guaranteed, and the robustness and the control precision of a control system under complicated industrial working conditions such as variable temperature and variable load are remarkably improved; and the requirements of high-end industrial equipment on high performance and high reliability of the motor driving system are met.

The invention discloses an AC excitation system rotor initial position on-line identification method and system, and the method comprises the steps: collecting a stator three-phase voltage signal when a stator is no-load, carrying out the Clarke conversion of the signal, obtaining a voltage component under a static coordinate system, and carrying out the low-pass filtering processing; calculating a stator flux linkage based on the stator current and the processed voltage component, and calculating a rotor flux linkage based on the rotor current; the rotor initial position offset is continuously accumulated through closed-loop adjustment by calculating the stator and rotor flux linkage angle and the phase difference until the phase difference approaches zero, and the accumulated rotor initial position offset is output; and combining the mechanical angle output by the encoder with the accumulated rotor initial position offset to obtain a corrected rotor position angle, and updating the corrected rotor position angle to an excitation system. According to the method, the in-phase characteristics of stator and rotor flux linkages are utilized, and the optimally designed low-pass filter is combined, so that the identification precision of the initial position of the rotor is remarkably improved, and the zero offset of the encoder is effectively compensated.

The invention relates to the technical field of lithium battery cell online nondestructive testing, and discloses a lithium battery cell lamination process online quality detection method, which is used for solving the problems of high difficulty in real-time online identification of hidden defects in a lithium battery cell lamination process and high omission ratio in a traditional method. The method comprises the following steps: firstly, synchronously acquiring an optical image of the surface of a pole piece and an X-ray perspective image of an internal structure on a lamination production line to obtain a multi-modal image sequence, carrying out time sequence registration and space alignment fusion on the two types of images, and extracting a pole piece layer, a diaphragm layer boundary and a density abnormal region; according to the method, an S-shaped deformation template and an internal foreign matter template are combined to carry out mode comparison, three-dimensional coordinate positioning, density grading and severity marking are carried out on an abnormal area, a structured abnormal catalogue is fed back to a production line control system, technological parameters are adjusted in a linkage mode, online nondestructive testing and closed-loop regulation and control of hidden defects are achieved, and the production efficiency is improved. The omission ratio is reduced; and the consistency and the safety of the lamination process are improved.

The invention relates to the technical field of servo drivers, in particular to a servo driver control method, and the technical scheme comprises the steps: carrying out the online identification of load inertia and external disturbance through a self-adaptive observer, generating a dynamic compensation parameter, constructing a multi-model parallel self-adaptive observer group, and carrying out the online identification of the load inertia and external disturbance through the self-adaptive observer group, a dynamic confidence evaluation mechanism is designed, disturbance dynamic compensation is realized, parameter sudden change and slow change scenes can be taken into consideration, the real-time performance of dynamic response can be ensured, compensation parameters are input into a fuzzy neural network controller, an optimal control quantity is generated in combination with a preset control target, a multi-source compensation parameter input channel is constructed, and the optimal control quantity is generated. A dynamic input weighting module is designed, fuzzy rule confidence is updated through an online strategy gradient algorithm to generate an optimized control quantity, the working condition adaptability can be enhanced, a dynamic target can be dealt with, finally, a sliding mode variable structure algorithm is adopted to carry out high-frequency buffeting suppression on the control quantity, and a final driving signal is output to a power module. And the control precision of the servo driver is improved.

The invention relates to the field of motor fault diagnosis, and particularly discloses a permanent magnet synchronous motor fault diagnosis method and system, which dynamically update motor physical model parameters through an online identification algorithm so as to reflect the characteristic change of a motor in real time. Thirdly, predicting the theoretical voltage of the healthy motor under the working condition by using the self-adaptive model and the real-time operation data, and calculating a residual sequence subjected to working condition normalization between the theoretical predicted voltage and the actual voltage; the residual signal is essentially free from the influence of working condition change, and only the abnormal characteristics caused by the fault are highlighted. And finally, inputting the high-robustness residual error sequence into a convolutional recurrent neural network, deeply fusing space-time fault features, and realizing accurate judgment on a motor fault type and confidence thereof, thereby effectively improving variable working condition adaptability and weak fault detection capability of diagnosis.

The application discloses an online error calibration method for a hemispherical resonator gyro based on frequency superposition, which solves the problem that the traditional error calibration method for a hemispherical resonator gyro cannot realize online calibration of gain error, bias angle error and phase error in the same working mode. The application utilizes high-frequency cosine and sine reference signals to demodulate two-channel detection signals and extract high-frequency components in the detection signals. Subsequently, an associated model is established by using the high-frequency cosine and sine given control signals and the demodulated signals to realize online identification of the gyro error, and the identification result is applied to online calibration of the channel gain error, bias angle error and phase error, thereby enhancing the anti-interference ability of the hemispherical resonator gyro under variable temperature conditions. The application is mainly used for simultaneous online calibration of the channel gain error, bias angle error and phase error.

The invention provides a digital twinning five-axis machine tool contour error prediction method and system and electronic equipment, and belongs to the technical field of numerical control machine tool precision control and digital twinning. According to the method, servo axis data are collected in real time at a numerical control system end, electromechanical parameters are identified on line, comprehensive disturbance is estimated, and an autoregressive exogenous (ARX) model is updated accordingly; an interpolation instruction is generated in a digital twin body end mirror image numerical control system process, an updated model is utilized to predict a shaft position with deviation, a tool path is obtained through kinematics positive solution, and finally contour errors are calculated through geometric alignment. According to the digital twinning five-axis machine tool contour error prediction method and system and the electronic equipment, dynamic and high-precision prediction of the five-axis machine tool machining track contour error is achieved through the online identification and digital twinning technology, the limitation of a traditional static compensation method is overcome, and the precision of the digital twinning five-axis machine tool contour error prediction is improved. And a key basis is provided for prospective process optimization and precision control, and the machining precision and efficiency are effectively improved.

The invention relates to a mechanical arm self-adaption compliance control method and system based on model reference self-adaption and impedance parameter online identification, and belongs to the field of robots. The method comprises the steps that a mechanical arm task space dynamics model is established; designing a reference model; a recursive least square method is adopted, and an equivalent stiffness matrix and a damping matrix of the interaction environment are identified online; designing a composite adaptive law; the expected rigidity of the reference model is dynamically adjusted according to the environment rigidity obtained through identification; and according to sensor data collected in real time, the mechanical arm task space dynamics model, the robot parameter self-adaption law, the environment rigidity, the environment damping and the environment parameter identification robust item, control force is generated, and the mechanical arm is driven through the control force. According to the method and the system, the environment characteristics can be sensed online, and the compliant behaviors can be adaptively adjusted, so that safe, efficient and high-robustness interaction of the mechanical arm in a complex and unknown environment is realized.

The invention relates to the technical field of cable defect identification, and discloses an incremental OS-ELM cable defect online identification method and system. The method comprises the steps of collecting cable joint monitoring signals to extract time-frequency features, calculating feature importance weighting initialization model weight, screening samples to adjust forgetting factor recursion update weight, constructing mixed batch optimization weight from a multi-scale time window, and fusing multi-classifier evaluation values to output a defect identification result. According to the invention, the accuracy and stability of cable defect online identification and the long-term knowledge retention capability are improved.

The invention relates to the technical field of copper bar machining equipment, and discloses a copper bar bending control system and device.The copper bar bending control method comprises the following control steps that S1, the initial temperature of a to-be-machined copper bar, the real-time temperature of a bending template and the actual pressure in a working cavity of a hydraulic cylinder are collected; s2, the actual pressure collected by the multi-source state sensing module is recorded to obtain pressure attenuation data, and the final accurate stress relaxation time constant of the current workpiece is calculated based on the pressure attenuation data; s3, generating a target pressure set value according to the final accurate stress relaxation time constant; s4, calculating in real time, and obtaining a control signal according to a deviation value between the generated target pressure set value and the actual pressure collected in S1; and S5, a control signal is received, and the pressure in the working cavity of the hydraulic cylinder is adjusted. According to the method, the springback difference caused by temperature fluctuation is effectively restrained through online identification and dynamic adjustment of the pressure maintaining strategy, and the product bending consistency and precision are improved.

The invention discloses a latex skinning defect online detection method and system based on machine vision, and relates to the technical field of machine vision, and the method comprises the following steps: obtaining image acquisition parameters, and constructing an industrial camera imaging model based on a latex production environment; based on the industrial camera imaging model, collecting target area image data of the latex surface; sequentially executing edge enhancement processing and connected region extraction operation on the target region image data to obtain a candidate defect region set; sequentially extracting geometric morphological features from the candidate defect region set, and screening suspected defect regions through a hierarchical judgment rule; and identifying a target defect area according to the number and spatial distribution characteristics of the suspected defect areas. The technical problem that the skinning defect in the latex production process is difficult to identify online can be effectively solved, and the method has good application prospects and popularization value.

The invention discloses an edge processing method, device and system for fluid system data, and the method comprises the steps: a dynamic data segmentation step: carrying out the dynamic segmentation of the operation data of the device based on a predefined or on-line recognized key event, the key event being an event representing the significant change of the operation logic or state of the device; when the occurrence of any key event is detected, taking the occurrence moment of the event as an end point of the current statistical aggregation period and a starting point of the next statistical aggregation period; a differential processing step: in each statistical period defined by the dynamic data segmentation step, performing differential statistical aggregation or feature extraction on different types of operation data according to a preset differential data processing strategy; and a data uploading step: packaging a result of the differentiation processing step into structured data, and uploading the structured data to a cloud server according to a preset uploading strategy. According to the method, the data transmission quantity is remarkably reduced, and meanwhile, the data analysis value is greatly improved.

The invention relates to a self-adaptive control method, system and device for environment test equipment, and the method comprises the following steps: S1, data collection: collecting environment detection data and data of a corresponding actuator according to a to-be-controlled physical quantity; s2, selecting or constructing a control object according to the physical quantity to be controlled; s3, performing online identification operation when an identification condition is met, otherwise, directly executing the step S4; s4, feedforward and feedback: calculating feedforward compensation and feedback compensation according to the value of the to-be-estimated quantity, and superposing the feedforward compensation and the feedback compensation to obtain a control instruction; and S5, performing constraint allocation on the actuator: performing saturation constraint processing on the control instruction to obtain a control quantity of the corresponding actuator. Joint online identification can be carried out on the to-be-estimated quantity only through perturbation excitation; different loads and test object states are automatically adapted; saturation constraint processing is adopted; the model and the algorithm are low in calculation amount, and precision, speed and engineering realizability are considered.

The invention provides a real-time monitoring system and method for process behaviors in an RH refining furnace. The system comprises an image acquisition and monitoring unit in the refining furnace based on an RH top lance, a data transmission unit, a data processing unit, a video display unit and a data display unit. Wherein the image acquisition and monitoring unit is used for acquiring real-time state information and image information in the RH refining furnace. And the real-time monitoring information is sent to the video display unit through the data transmission unit, and real-time pictures in the refining furnace are displayed. The data display unit displays the real-time refining process, the real-time refining temperature, the existence of top lance flames, the top lance height, the molten steel liquid level height and other information. According to the method, real-time online identification and prompt in the RH refining process can be effectively realized, and good support is provided for improving refining safety, refining intelligence, equipment integration, reduction of use of disposable temperature measurement equipment, judgment and improvement of equipment faults, refining process optimization and refining efficiency optimization.

The invention discloses a load self-adaption device based on online system identification. The device comprises a track-Jacobian matrix offline generation module used for obtaining a pseudo-inverse matrix of a track-Jacobian matrix offline; the online excitation and measurement module is used for online controlling the robot to execute excitation motion and measuring to obtain an actual joint response error vector; and the rapid parameter identification module is used for combining the actual joint response error vector with the pseudo-inverse matrix to obtain an inertial parameter of a target load. The track-Jacobian matrix represents a linear sensitivity relation between inertia parameter disturbance of a load at the tail end of the robot and a joint response error. According to the method, the dynamic solving process with high calculation complexity is transferred to the offline stage to be completed, and the deterministic matrix-vector operation only needs to be executed once in the online stage, so that the calculation delay of online identification is greatly reduced on the premise of ensuring the identification precision.

The invention discloses a stainless steel welded pipe drawing process regulation and control method and system based on online identification, and the method comprises the steps: synchronously collecting a temperature field distribution image, axial and circumferential strain time sequence signals and a microscopic plastic deformation acoustic emission signal of the surface of a welded pipe, and generating a time-space synchronous multi-mode perception data flow; performing cross-domain feature fusion processing on the multi-modal sensing data stream, and extracting a fusion feature vector; inputting the fusion feature vector and the real-time drawing process parameters into an online rolling identification module, and outputting a flow stress curve and an anisotropy coefficient; based on the flow stress curve and the anisotropy coefficient, generating a cooperative regulation and control instruction set; and executing the coordinated regulation and control instruction set, and driving the drawing machine main motor, the mold posture adjusting mechanism and the minimal quantity lubrication system to perform linkage adjustment. By utilizing the embodiment of the invention, the self-adaptive closed-loop optimization regulation and control in the drawing process of the stainless steel welded pipe can be realized, and the product quality, the process stability and the production efficiency are improved.

The invention discloses a motor multi-parameter online identification and robust control method based on multi-observer cascade. The method comprises the steps of obtaining real-time operation data of the permanent magnet synchronous motor; obtaining a preliminary observation value through a sliding mode observer group; the sliding-mode observer group is constructed based on a mathematical model of the permanent magnet synchronous motor and is used for receiving real-time operation data and synchronously outputting initial observation values of a permanent magnet flux linkage, q-axis inductance and d-axis inductance; then analyzing the preliminary observation value to obtain a flux linkage accurate value, a q-axis inductance accurate value and a d-axis inductance accurate value, and finally performing adaptive control on the permanent magnet synchronous motor according to the flux linkage accurate value, the q-axis inductance accurate value and the d-axis inductance accurate value. According to the method, multiple motor parameters can be synchronously estimated, the limitation that other parameters are assumed to be accurate when a single parameter is identified in a traditional identification method is avoided, adaptive current reference and voltage decoupling are further generated based on the estimated parameters, and robust and stable field weakening control under variable working conditions can be achieved.

The invention belongs to the technical field of power system stability analysis and control, and relates to a wind power plant characteristic root rapid solving method and system based on the circular ring theorem, and the method comprises the following steps: S1, carrying out the online identification of impedance; s2, constructing a wind power plant grid-connected system characteristic equation; s3, establishing a circular ring theorem model; s4, analyzing the stability of the wind power plant grid-connected system; according to the technical scheme, the calculation method of the circular ring theorem is adopted, the order number of the wind power plant grid-connected system does not need to be reduced, the feature root of the high-order wind power plant grid-connected system can be rapidly calculated, the solving efficiency is high, meanwhile, the directional capturing capacity of the main feature root is high, and the main feature root of the wind power plant grid-connected system can be rapidly searched.

The invention discloses a real-time energy consumption optimization control system for digital twins of a paper cup production line, and relates to the technical field of energy consumption optimization control, and the system comprises the steps: 1, collecting incoming material and environment fingerprints, carrying out the semantic alignment of an asset management shell, and building an energy consumption baseline; 2, constructing a mass-energy coupling twinborn model, and fusing infrared, vision, output intensity, risk and unit energy consumption; 3, online identification is carried out, fingerprints are mapped into process boundaries and change rates, and a process window is published; 4, joint optimization is carried out in the window, and reinforcement learning of shielding projection is used for short-time takeover during disturbance; and 5, safely issuing, storing the strategy snapshot, auditing and rectifying to form a closed loop. Multi-source connection, energy conservation, consumption reduction, quality control, peak power suppression and traceable treatment are realized; and compressed air supply-transmission and distribution-gas utilization efficiency evaluation, boundary versioning and rate limitation unified management, boundary-crossing back-off of a safety template and strategy parameter closed-loop updating are carried out.

The invention relates to the technical field of power system stability analysis, and provides a broadband oscillation modal online identification method, which comprises the following steps: constructing a dynamic heterogeneous graph model according to the physical topology and measurement data of a power system, the dynamic heterogeneous graph model comprising a node set, an edge set and dynamic attributes; inputting the dynamic heterogeneous graph model into a physical information embedded space-time graph convolution circulation network, extracting space-time features and outputting potential states; based on the potential state and a learnable state matrix in the network, broadband oscillation modal parameters are calculated, and the broadband oscillation modal parameters comprise oscillation frequency, a damping ratio and a participation factor; and according to the real-time measurement data, carrying out online updating on the space-time diagram convolution circulation network embedded with the physical information so as to maintain the identification precision and enable the closed-loop operation of the digital twin model of the power system. And furthermore, the problem of online identification of the broadband oscillation mode of the high-proportion new energy power grid is solved.

The invention provides an MRAS permanent magnet synchronous motor parameter online identification method based on BP neural network optimization, and the method comprises the steps: constructing two BP neural network models, carrying out the offline training optimization of the two models, and obtaining an optimal neural network model which can accurately predict the flux linkage of a rotor and the inductance of a stator; based on an MRAS step-by-step identification strategy, two-stage motor parameter step-by-step identification units are built, the two optimal neural network models are embedded into the corresponding motor parameter step-by-step identification units, and based on the problem that the stator resistance identification effect of the BP neural network is poor, the two optimal neural network models are embedded into the corresponding motor parameter step-by-step identification units; according to the invention, the BP neural network and the adaptive rate unit are established in the second MRAS parameter identification unit, and the online dynamic identification of the stator resistance and the stator inductance is realized through the combination of the BP neural network and the adaptive rate unit. According to the method, an off-line training-on-line identification mode is adopted, the complex parameter setting process in a traditional method is avoided, the identification efficiency is remarkably improved, and the identification result is accurate.

The invention relates to the technical field of traffic safety research, in particular to a vehicle abnormal driving behavior recognition method and recognition system based on deep learning. The method comprises the steps of obtaining historical driving data and station environment data to construct an offline training data set, extracting a first type of time sequence features representing abnormal driving behaviors and a second type of time sequence features representing abnormal scenes, determining an optimal feature combination based on neural network learning, outputting a first feature vector representing the abnormal driving behaviors, and outputting a second feature vector representing the abnormal driving behaviors. And according to the first feature vector representing the abnormal scene and the second feature vector representing the abnormal scene, establishing an abnormal driving recognition model, inputting the driving data and the scene data acquired in real time into the abnormal driving recognition model, and outputting to obtain an abnormal driving behavior recognition result of the target operation vehicle. According to the invention, a dual-channel feature training mechanism is adopted, millisecond-level online identification is realized based on offline training, and a vehicle abnormal behavior result is output in real time.

The invention discloses a wear online identification and compensation control system and method for a high-pressure distribution plunger and barrel assembly, and the system comprises a data collection module which is used for collecting working condition data and state data; a health state dynamic baseline model obtained through offline training is arranged in the data processing module, and the wear level of the plunger and barrel assembly is recognized by calculating the residual error between actual state data and health state data predicted by the model under the current working condition; and the compensation control module is used for generating a self-adaptive compensation control instruction according to the identified wear grade and the current working condition and sending the self-adaptive compensation control instruction to an execution mechanism. According to the method, the dynamic base line synchronous with the real-time working condition is constructed, early weak wear characteristics can be separated from strong working condition interference, the recognition sensitivity and accuracy are improved, the influence of wear on the system performance is actively counteracted through closed-loop adaptive compensation control, the service life of equipment is expected to be prolonged, and the system performance is improved. And technical support is provided for predictive wear monitoring management.

The invention discloses an industrial mechanical arm joint rigidity online identification method based on an optimal excitation track, and belongs to the field of mechanical arm parameter identification. The method comprises the following steps: firstly, establishing a mechanical arm rigid-flexible coupling dynamic model, and constructing a linear regression equation taking joint rigidity and viscous damping as parameters; a Fisher information matrix is derived based on the regression model to quantify the amount of information of the excitation trajectory. And further, parameterizing the excitation trajectory by using finite term Fourier series, and solving the optimal excitation trajectory by using a scalar index of a Fisher information matrix and a sustainable excitation condition as optimization targets under the condition of meeting joint motion constraints. And finally, the mechanical arm is driven to execute the track, motion data are collected and input into parameter identification algorithms such as unscented Kalman filtering and recursive least square, and online and high-precision estimation of the time-varying joint stiffness is achieved. According to the method, the excitation track is designed directly from the information theory, and the precision, robustness and real-time performance of joint rigidity identification are effectively improved.

The invention discloses an AI vision intelligent inspection and foreign matter recognition emergency shutdown system for a belt conveyor. The system comprises data acquisition and sample construction; constructing a DETR foreign matter recognition model; images are collected in real time, and online recognition is conducted through a DETR foreign matter recognition model; carrying out target tracking and behavior analysis based on a multi-target tracking algorithm of ByteTrack; a distance threshold value, a speed threshold value and a score threshold value are set; graded alarming is carried out, and a shutdown signal is sent to the PLC control unit under the shutdown condition. The system comprises an alarm module, an edge calculation unit, a communication module and a PLC control unit. The AI inspection camera, the edge calculation unit, the communication module and the PLC control unit are connected in sequence; the edge calculation unit comprises an AI identification module and a tracking and behavior analysis module. According to the invention, real-time identification, behavior judgment and automatic safe shutdown linkage control of foreign matters or abnormal states on the conveying line can be realized.

The application discloses a kind of porous metal deformation mechanism acoustic emission online identification method, device and electronic equipment, including collection porous metal material acoustic emission signal and characteristic parameter data and pre-processing, obtain each acoustic emission signal feature vector;Determine the loss function of porous metal material deformation, build acoustic emission machine learning classifier candidate model;Train multiple different parameters machine learning classifier candidate model;Evaluation and optimization are carried out, and preferred machine learning classifier model is selected;Preferred machine learning classifier model is applied to online identification of the deformation mechanism of porous metal material acoustic emission signal.This method can be combined with the loss function determined based on the distribution characteristics of acoustic emission signal of different deformation mechanisms, build machine learning classifier model, have the characteristics of not needing label data and high identification accuracy and effectiveness.