115 results about "Potentially abnormal" patented technology

The invention discloses a large and small model-based data anomaly detection method, system and device and a medium, belongs to the technical field of data processing and artificial intelligence, and aims to solve the technical problem of how to combine the rapid screening capability of a small model and the deep analysis capability of a large model to realize efficient and collaborative anomaly detection and cleaning. According to the technical scheme, an improved Isolation Forest small model is used for rapid screening, potential anomalies in large-scale data are rapidly detected through a dynamic threshold adjustment mechanism, adaptive feature weight distribution and multi-scale division optimization, an anomaly score is generated, and suspected abnormal data are preliminarily screened; secondly, performing fine detection on suspected abnormal data by utilizing a large model, and accurately identifying an abnormal type and generating a high-reliability abnormal result through deep semantic embedding analysis, abnormal feature clustering and domain rule verification; and finally detected abnormal data are classified and labeled, and the data characteristics are combined to process the abnormal data by adopting a deletion, correction or marking cleaning strategy.

The invention relates to the technical field of motor testing, in particular to an online monitoring method and system for the health state of a carbon brush of a motor, and the method comprises the steps: determining a target operation load environment cluster of a target carbon brush at the current moment according to the pre-obtained load condition and operation environment condition of the target carbon brush in the current operation time period; screening out a reference operation time period; determining an abnormal loss factor and a potential abnormal wear feature vector of the target carbon brush at the current moment; according to the Lyapunov index of the vibration signal of the target carbon brush in the current operation time period and the temperature change and the current change of the target carbon brush in the current operation time period, determining a wear fault display possible feature vector; therefore, the health state of the target carbon brush at the current moment is determined. According to the invention, the potential abnormal wear condition of the target carbon brush at the current moment is considered, the health state of the carbon brush is monitored, and the timeliness of abnormal health state monitoring of the carbon brush is improved.

The invention relates to an electric power internet of things operation and maintenance management method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining sensor data, carrying out the cleaning and time alignment of the sensor data, and obtaining standard operation data; based on a historical data window corresponding to the standard operation data, calculating a mean value and a standard deviation of the historical data, and further generating a dynamic detection threshold value; comparing the standard operation data of the current time node with a dynamic detection threshold value, and screening out potential abnormal data; calculating an abnormal trend factor corresponding to the potential abnormal data, and determining an abnormal development trend of the potential abnormal data according to the abnormal trend factor; and determining a corresponding causal influence weight based on a preset causal relationship network, and calculating a fault score according to the causal influence weight and the abnormal trend factor, so as to generate an early warning report according to the fault score. The method has the effect of improving the management efficiency of the power equipment.

The invention discloses a network equipment abnormity monitoring and early warning method and system based on AI, and belongs to the technical field of network equipment monitoring, and the system module comprises a multi-source data acquisition module which obtains various characteristic parameters of network equipment; the data transmission and processing module is used for transmitting and preprocessing the characteristic parameters; the intelligent analysis module performs comprehensive analysis based on the characteristic parameters and judges whether the network equipment is abnormal or not; and the early warning response module performs hierarchical response based on a judgment result of the parameter analysis module. According to the invention, under the normal operation condition of the network equipment, the potential abnormal risk of the network equipment can be judged according to the influence condition of the environment where the network equipment is located and the relative deviation condition of each piece of network equipment, so that the potential abnormal risk can be handled in time, thereby reducing the occurrence of subsequent abnormal situations and improving the safety of the network equipment. And normal and stable operation of network equipment is ensured.

The invention discloses an intelligent data anomaly detection method and system based on an Internet of Things integrated management and control technology, and the method comprises the steps: S1, collecting and preprocessing intelligent data, and constructing a standardized data set; s2, performing preliminary anomaly detection on the standardized data set by adopting an isolated forest algorithm, and screening out potential abnormal data points; s3, constructing a self-supervised training sample, performing enhancement detection on the abnormal data point set, and generating a correction factor; s4, recalculating an abnormal score for each abnormal data point based on the correction factor; s5, abnormal data points output through re-correction are classified and stored, and log information is generated; and S6, dynamically optimizing the model parameters according to a detection result fed back in real time. According to the method, an efficient and scientific optimization scheme can be provided in Internet of Things intelligent data anomaly detection, and remarkable technical values and economic benefits are brought to practical application.

The invention discloses an elevator operation fault early warning method and system, relates to the technical field of intelligent elevator control, and aims to improve the operation safety and fault response capability of an elevator in a potential abnormal state. By constructing a load critical drift test working condition, a sudden non-vertical stress test working condition and a weak power supply fluctuation interference test working condition, abnormal behaviors which are difficult to appear but have potential safety hazards in the normal operation process of an elevator are induced; the method comprises the following steps: acquiring operation data under various working conditions, respectively calculating a drift overload response coefficient, a guide rail non-vertical interference coefficient and a power supply induced instability coefficient, respectively comparing the coefficients with preset thresholds, and judging whether operation faults such as brake response mismatching, asynchronous action of safety tongs and door lock control interference exist or not; and if the judgment result is abnormal, a targeted strategy is automatically generated and early warning is given out. According to the method, active recognition and classified response of elevator faults can be achieved, and the maintenance efficiency and the operation reliability are effectively improved.

The invention discloses a grotto temple cavern microenvironment monitoring data quality control and evaluation method, and particularly relates to the field of grotto temple cavern microenvironment monitoring data, and the method comprises the steps: obtaining grotto microenvironment monitoring data, carrying out the time-space reconstruction of the grotto microenvironment monitoring data, and building a data expression basis with a consistent attribute structure. An execution entry is provided for subsequent processing; potential abnormal fragments are identified from the cross relation between the time rhythm and the numerical value change, abnormal points are eliminated, and the validity of the cave microenvironment monitoring data value domain is guaranteed; the problem of disordered timestamp sequence is recognized, a correct sorting structure is recovered in combination with a trend rule, and time sequence consistency is guaranteed. By constructing a multi-stage closed-loop processing chain with rhythm change recognition, numerical mutation elimination, trend sequence recovery and structure quality fusion as the core, structure unification, anomaly recognition, time sequence correction and credibility output of monitoring data are achieved, and therefore effectiveness, consistency and application value of the monitoring data are systematically improved.

The invention relates to an abnormal data identification method and device, computer equipment and a storage medium. The method comprises the following steps: constructing a user social network graph according to user communication information; dividing risk groups according to the user login information and the user transaction information to obtain high-risk groups; performing association marking in the user social network graph according to the high-risk group to obtain a potential abnormal group; marking a target user to be identified for the user login information and the user account information of the potential abnormal group; and obtaining the logout times of the to-be-identified target user, and marking the to-be-identified target user as target abnormal data according to the logout times. The method and the device can be applied to financial anti-fraud application scenes, target abnormal data with fraud risks in users can be effectively and accurately identified, and the identification reliability is improved.

The invention provides a data evaluation method and device based on historical local anomalies, equipment and a medium. The method comprises the following steps: performing data preprocessing on historical data in environmental monitoring, calculating a k-neighborhood distance, a reachable distance, a local reachable density and a local anomaly factor, and generating reference density distribution information; processing the reference density distribution information, merging test data points with historical data one by one to calculate an LOF value, screening out potential abnormal samples in combination with a set threshold, and generating potential abnormal sample information; for potential abnormal samples, Z-Score is calculated according to feature dimensions, and abnormal index information of each dimension is generated in combination with a set threshold value; processing the potential abnormal sample information and the abnormal index information of each dimension, marking a sample abnormal state and positioning an abnormal source index, and generating a sample abnormal state mark and abnormal source index positioning information; and processing the sample abnormal state mark and the abnormal source index positioning information to generate abnormal information of the environmental data.

The invention relates to the technical field of data analysis, and discloses a trajectory prediction and abnormal behavior detection system and method, and the method comprises the steps: obtaining a target monitoring image; constructing a target track segment of the moving target based on the target monitoring image; performing prediction completion on the target trajectory segment to obtain a target trajectory of the moving target; calculating the compression degree and the expected speed deviation of the target trajectory; identifying a potential anomaly of the target trajectory based on the degree of compression and an expected velocity deviation; if the target trajectory has potential abnormality, predicting a target expected trajectory based on the target monitoring image; and calculating a matching degree between a target trajectory and the target expected trajectory, and identifying an abnormal target trajectory based on the matching degree. According to the invention, the abnormal track of the moving target can be accurately identified, and the efficiency and the foresight of abnormal track detection are improved.

The invention discloses a control method and device of a fault diagnosis system and a storage medium, and belongs to the technical field of fault diagnosis systems. The method comprises the steps that fault description information and multimedia data are received, text features and multimedia features in the fault description information and the multimedia data are extracted, the text features and the multimedia features are fused based on an attention fusion module, fusion features are generated, sensor data of to-be-analyzed equipment are obtained, sensor time sequence features are generated, and the fault description information and the multimedia data are analyzed; and the fusion features and the sensor time sequence features are mapped to a hidden space, the fusion features and the sensor time sequence features are aligned in the hidden space, joint representation is generated, and a target fault type is identified through a fault classification model. According to the method and the device, the user subjective description and the joint representation of the equipment objective signal are fused, so that the joint representation can simultaneously capture the fault phenomenon details described by the user and the potential abnormal mode in the sensor time sequence signal through cross-modal feature coupling, and the comprehensiveness and the reliability of the fault diagnosis result are improved.

The invention provides an electroencephalogram abnormal signal detection method based on step-by-step identification and multi-agent decision, and relates to the technical field of biomedical signal processing.The method comprises the steps that electroencephalogram signal data of an epilepsy patient are collected, and the data are analyzed according to a preset channel sequence and cut into a plurality of signal segments with the same length; extracting a multi-dimensional feature from each signal segment; inputting the time domain feature, the frequency domain feature and the inter-channel synchronization feature of each signal segment into an isolated forest model, and screening based on an abnormal proportion threshold to obtain at least one potential abnormal segment; based on the depth scattering feature, the wavelet transform feature, the time domain feature and the frequency domain feature of each potential abnormal segment, executing a multi-agent integration decision on each potential abnormal segment to obtain a comprehensive abnormal score corresponding to each potential abnormal segment; and determining an abnormal signal in the electroencephalogram signal data based on the comprehensive abnormal score corresponding to each potential abnormal segment. According to the method, the false alarm rate of electroencephalogram abnormal signal detection is remarkably reduced.

The invention provides an Internet of Things data acquisition and abnormity early warning method and system, and relates to the technical field of Internet of Things data processing, and the method comprises the steps: constructing a digital twin model for a physical entity at a cloud end, and issuing prediction data to an edge node; acquiring sensor data by an edge computing node at a basic frequency, and computing a residual error between real-time data and predicted data and an information entropy of a residual error sequence; dynamically adjusting the data acquisition frequency of the edge node based on a two-dimensional decision space formed by the residual error and the information entropy; and when the residual error and the information entropy both exceed the threshold values, potential abnormity is judged, an edge side abnormity analysis program is triggered, and abnormal event information is uploaded. According to the invention, through cloud edge cooperation and intelligent decision making, the acquisition frequency is reduced in a stationary period, resources are saved, the frequency is improved when an abnormal symptom appears, monitoring is enhanced, and optimal balance between resource consumption and monitoring precision is realized; double criteria are adopted to effectively filter noise interference, and the false alarm rate and the missing report rate are reduced.

The invention relates to the technical field of wastewater pipelines, in particular to a health state diagnosis method for an electrochemical energy storage prefabricated cabin wastewater collection pipeline, which comprises the following steps: collecting an operation signal of the pipeline, simulating the operation state of the pipeline based on a preset simulation model, obtaining an operation virtual signal of the pipeline, comparing the operation signal with the operation virtual signal, and determining the health state of the pipeline. Operation deviation is obtained; extracting the characteristics of the operation deviation to obtain a deviation characteristic vector; and inputting the deviation feature vector into a fault classification model obtained by training, identifying a potential abnormal type of the pipeline, and outputting a health state index. By collecting flow, sound and vibration signals, generating operation virtual signals in combination with a simulation model, comparing and analyzing deviation characteristics, inputting the deviation characteristics into a fault classification model, identifying potential anomalies and outputting health state indexes, quantitative evaluation and early warning of pipeline health are achieved, the pipeline operation state can be reflected in multiple dimensions, and the reliability of the system is improved. Diagnosis accuracy and system reliability are improved, and health grade division and operation and maintenance decision making are supported.

The invention discloses a threat intelligence automatic generation method and device. The method comprises the following steps: acquiring safety behavior data in real time; performing feature extraction on the security behavior data to obtain target behavior features; comparing the target behavior characteristics with a boundary threshold range of a pre-trained normal baseline model; the normal baseline model is obtained based on the boundary threshold range of the behavior feature data obtained by feature extraction; inputting the target behavior characteristics into a pre-trained anomaly detection model to obtain a current anomaly score; the anomaly detection model is obtained by constructing a random isolation tree based on a historical feature matrix obtained based on historical behavior features and training based on the random isolation tree; if the current exception score is greater than a preset exception sample threshold value, judging the security behavior data as first exception data; and the first abnormal data is processed to obtain threat intelligence, so that the potential abnormal behaviors are accurately identified, and the threat detection efficiency and response capability in a complex environment are improved.

The invention relates to the field of vulnerability mining, in particular to a vulnerability mining method based on a large model, and the method comprises the steps: constructing a label combination library for a working condition terminal, and subsequently constructing a correlation model for a label combination based on a credible log fragment corresponding to each label in the label combination; based on the occurrence probability of each label combination and combined with the constraint discrete features of the association model corresponding to the label combination, effective sub-labels are set for the label combination, a current label combination of a current log record is analyzed subsequently for a new log record, and adaptability analysis is performed on the log record according to the effective sub-labels corresponding to the current label combination, so that the log record is obtained. According to the method, potential laws of the operation process of the industrial control terminal are considered, the association model of the log records under the specific label combination is constructed, the log records are analyzed based on the effectiveness adaptability of the label combination, log fragments possibly representing that potential abnormal vulnerabilities exist in a system are rapidly marked when massive log records are faced, and the system reliability is improved. And the accuracy and efficiency of vulnerability mining are improved.

The invention relates to the field of big data analysis, in particular to a prediction method and system based on industrial big data. The method comprises the following steps: extracting an equipment data stream in a factory production process, and carrying out multi-cycle division and centralized visualization to generate a visual production data stream; performing deviation evaluation according to the visual production data flow, and marking potential abnormal equipment; calculating the change rate and the change amplitude of the potential abnormal equipment; positioning fault equipment according to the change rate and the change amplitude, and marking the fault equipment; performing multi-parameter linkage analysis and deviation attribution analysis on the fault equipment, and identifying equipment fault factors; and performing equipment failure shutdown prediction and equipment maintenance priority planning based on the equipment failure factors, and generating an equipment maintenance sequence. Based on factory big data analysis, the equipment fault time point is accurately predicted, and the accuracy of fault diagnosis and the shutdown maintenance efficiency are improved.

The invention relates to the field of project management, in particular to a data analysis method for project management, and the method comprises the steps: obtaining the high-dimensional data of a scaffold after preprocessing, the high-dimensional data comprising the bearing capacity, the inclination angle and the space coordinates; classifying the high-dimensional data of the scaffold into a plurality of clusters by utilizing an iterative self-organizing clustering algorithm; wherein in the clustering process, the inter-cluster separation degree and the intra-cluster compactness of each clustering cluster are calculated, and if the mean value of the inter-cluster separation degree and the intra-cluster compactness of the current iteration is larger than that of the last iteration, the number of iterations is increased; otherwise, reducing the number of iterations; and performing anomaly detection on the high-dimensional data in each cluster, and when the high-dimensional data of the scaffold at the same position is continuously marked as abnormal data points for multiple times, judging that the position is abnormal. According to the method, the clustering result is dynamically optimized, so that the potential abnormal area in the scaffold structure can be effectively identified, and the safety and the reliability of the scaffold are improved.

The invention discloses a bone nail automatic identification and dosage confirmation method based on multi-feature collaborative reasoning. The method comprises the following steps: firstly, collecting a multi-view image sequence of a bone nail box in an initial state to obtain a bone nail recognition image sequence; based on the image sequence, extracting boundary information of a bone nail placing position, and generating a bone nail structure map used for representing a relative position relation of placing areas; combining the structural atlas with the image sequence, extracting image features of each placement area, and performing comparison identification based on standard shape features of the bone nail to obtain an identification result; the recognition result is compared with a historical use behavior mode, potential abnormity is recognized, and a corrected enhanced recognition result is output; respectively executing the steps before and after an operation, obtaining an enhanced recognition result and performing difference comparison to obtain bone nail use difference information; and bone nail use confirmation information is generated in combination with the operation identifier, the identification confidence and the acquisition time, and is written into a non-tampering storage medium, so that closed-loop tracing of nail use records is realized.

The invention relates to the field of deep learning, and discloses a heart failure patient home care intelligent monitoring method combined with deep learning, which comprises the following steps: continuously acquiring the heart rate, blood pressure, breathing and body movement states of a patient in a home environment; performing conjoint analysis of time and activity states on the data streams after priority sorting, and forming a dynamic health feature set in combination with historical health records of patients and environment variable information; comprehensively evaluating the dynamic health feature set by using a deep learning model, and identifying a potential abnormal mode of short-time change through the deep learning model; marking and grouping the physiological parameters and related behavior scenes corresponding to the candidate abnormal events, locally evaluating the current health state of the patient based on grouping information, and identifying a potential high-risk condition; and according to the local evaluation result and the overall health trend of the patient, generating real-time early warning and personalized nursing suggestions. The method has the advantage of improving the identification capability of short-time abnormity.

The application relates to the technical field of network security, in particular to an encryption scene detection method and system based on double-model adaptive switching, which comprises the following steps: loading an unsupervised detection model and an unsupervised data collection module; the unsupervised data collection module collects data for the unsupervised detection model to perform unsupervised preliminary screening and obtain a real-time anomaly score; an adaptive switching controller compares the real-time anomaly score with a switching threshold to determine whether the system is abnormal; if the system is normal, the unsupervised preliminary screening is continued; if the system is abnormal, supervised fine screening is performed; a supervised data collection module collects behavior data and inputs the behavior data into the supervised detection model to determine whether the attack is an encryption virus attack; if the attack is an encryption virus attack, corresponding strategies are executed; if the attack is not an encryption virus attack, the unsupervised preliminary screening is continued. The application performs coarse-grained threat preliminary screening in a normal state, and adaptively switches the model to perform fine screening when potential abnormalities are detected, thereby guaranteeing a high detection rate and reducing the influence on system performance.

The application discloses a digital thread driven intelligent management system for a clinical laboratory, which comprises a system body, and the system body is operated through the following method, and the specific method comprises the following steps: obtaining patient historical records and same batch sample data from a pre-established database by collecting current test result data, and obtaining multimodal information containing original images, detection curves and trend deviations; determining potential abnormal patterns by processing the original images and the detection curves through an abnormality detection algorithm according to the multimodal information; if the abnormal patterns are inconsistent with historical comparison results, comparing previous results of the patient through a trend analysis algorithm to judge the trend deviation degree; obtaining critical project specific threshold values according to the trend deviation degree to obtain key clue recognition results; and the application aims to solve the problem that in the prior art, when facing critical value projects, comprehensive verification and rapid release cannot be balanced, resulting in poor quality control and efficiency of the clinical laboratory.

The invention relates to the technical field of medical data processing, in particular to a urinary post-operation sign monitoring data processing method and system, and the method comprises the steps: obtaining original urinary post-operation sign time sequence flows of a plurality of monitoring devices, and carrying out the verification of data integrity and consistency. And creating a dynamic sign map for describing physiological state evolution, and injecting sign nodes and state transition edges into the dynamic sign map based on the verified time sequence flow. And performing staged partitioning on the dynamic sign map according to a postoperative rehabilitation stage division standard. And deploying independent abnormal probes matched with the stage characteristics in each partition, and performing parallel operation to scan and collect potential abnormal signals. And constructing an abnormal signal aggregation network, and carrying out fusion and priority ranking on the collected abnormal signals. And according to the network output, generating a differentiated graded early warning instruction, and calling a corresponding early warning response protocol for execution. According to the method, modeling of continuous evolution of the postoperative physiological status and staged adaptive anomaly monitoring are realized.

The invention provides a piezoelectric monitoring system and method for internal damage of a concrete pile foundation, and the method comprises the steps: calibrating a potential abnormal region of the concrete pile foundation according to first detection data generated by all piezoelectric intelligent aggregates in the concrete pile foundation in a first time period, thereby screening key piezoelectric intelligent aggregates; the monitoring range of the concrete pile foundation is effectively reduced; according to second detection data generated by all the key piezoelectric intelligent aggregates in the second time period, stress distribution change characteristics of the potential abnormal area are obtained, so that the damage change trend of the potential abnormal area is estimated, and accurate structural damage identification of the concrete pile foundation is achieved; and according to the damage change trend, expected deformation characteristics of the potential abnormal area are determined, so that the overall construction state of the concrete pile foundation is predicted, a damage early warning message is generated, the potential structural damage condition of the concrete pile foundation is effectively predicted, and the structural monitoring efficiency and reliability of the concrete pile foundation are improved.

The embodiment of the present application discloses an electrocardiogram signal processing method, apparatus, device and storage medium, which belongs to the field of information processing technology. In the embodiment of the present application, a plurality of cardiac signal segments are first intelligently classified by a classification model, so as to find out the potential abnormal ST segments in each cardiac signal segment. Afterwards, the voltage value and the corresponding baseline voltage value of the potential abnormal ST segment are determined according to the classification result of each cardiac signal segment, and then based on the voltage value and the corresponding baseline voltage value of the potential abnormal ST segment, it is further determined whether the potential abnormal ST segment is a truly abnormal ST segment. In this way, not only can the accuracy of ST segment abnormality detection be guaranteed and the probability of missed detection be reduced, but also the efficiency is higher than manually searching for abnormal ST segments from electrocardiogram signals.

This invention provides a method for detecting abnormal signals in electroencephalograms (EEGs) based on step-by-step identification and multi-agent decision-making, belonging to the field of biomedical signal processing technology. The method includes: acquiring EEG signal data from epilepsy patients; parsing the data according to a preset channel order and segmenting it into multiple signal segments of equal length; extracting multi-dimensional features from each signal segment; inputting the temporal, frequency, and inter-channel synchronization features of each signal segment into an isolated forest model, and filtering for at least one potential abnormal segment based on an abnormality ratio threshold; performing multi-agent integrated decision-making on each potential abnormal segment based on its depth scattering, wavelet transform, temporal, and frequency features to obtain a comprehensive abnormality score corresponding to each potential abnormal segment; and determining the abnormal signal in the EEG signal data based on the comprehensive abnormality score corresponding to each potential abnormal segment. This invention significantly reduces the false alarm rate of abnormal signal detection in EEGs.

Techniques for remotely monitoring a patient and a corresponding medical device are described. The remote monitoring includes identifying a first set of images representing a particular location of a patient's body in which at least one component of an implantable medical device (IMD) coincides, determining a projection of a change characteristic of the particular location of the body, identifying a second set of images, determining a second set of change characteristics, comparing the second set of change characteristics to the projection, and identifying a potential abnormality at the particular location of the body.

This invention discloses a data analysis-based method and system for early warning of anomalies in basic nickel carbonate production. The method includes: acquiring historical monitoring data within a preset time period; selecting key target data points by calculating data deviation; and extracting multiple historical monitoring data sequences with adjacent target data points as start and end points. During real-time monitoring, when the current monitoring data does not exceed a preset threshold, it is matched with the most similar historical sequence. By analyzing the subsequent data change rate trend of the sequence, it is determined whether there are potential anomalies in the current data. Combining static threshold early warning with dynamic trend early warning can identify risks that are still within the safe range but have abnormal development trends, realizing the transformation from post-event alarm to pre-event early warning. This significantly improves the timeliness and accuracy of early warning in the basic nickel carbonate production process, effectively ensuring production safety and stability.

This invention relates to the field of big data storage technology, specifically to a big data storage service method and system. The method analyzes the temporal growth trend of ad views within the current period, combining historical comparative analysis and the stability of view growth to screen for potentially abnormal periods. Through correlation analysis of page jump sequences within these suspected abnormal periods and their differences from sequences in non-abnormal periods, it obtains abnormal jump indicators. Combined with abnormal user behavior within the current period, it determines normal access behavior. Based on normal access behavior, it integrates ad access and effective interaction data from each user to obtain storage priority for each ad and optimize resource allocation. This invention identifies normal access based on temporal behavior and jump logic differences, and dynamically adjusts ad storage strategies based on user interaction depth, reducing resource waste caused by malicious behavior and improving the effectiveness of storage resource utilization.

The invention relates to the technical field of coal mining, and discloses a coal mining-based driving working face abnormal information detection method and system, and the method comprises the steps: obtaining an acoustic emission signal generated when a heading machine works on a driving working face, separating a cutting pick-coal rock impact signal and a coal rock mass fracture signal based on a propagation path and waveform characteristics of the acoustic emission signal; identifying the wear state of the cutting pick based on the frequency attenuation characteristic of the cutting pick-coal rock impact signal, and extracting a coal rock fracture dominant frequency band based on the spectrum mutation characteristic of the coal rock mass fracture signal; compensating the energy of the coal rock fracture dominant frequency band according to the wear state of the cutting pick, and establishing a co-evolution relationship between the compensated energy and the spectrum mutation characteristic in a time sequence; according to the method, the energy precursor abnormal fluctuation is sensitively captured, so that early-stage, accurate and targeted detection of potential abnormal information of the driving working face is realized, and the accuracy and timeliness of early warning are remarkably improved.