73results about How to "Enhanced description ability" patented technology

This invention discloses a knowledge- and data-driven time series forecasting method and system, belonging to the field of industrial time series forecasting technology. To address the technical problems of existing technologies when processing strongly non-stationary data, such as key signal attention dilution, systematic underestimation of peak amplitude, incomplete capture of multi-scale dynamic patterns, and difficulty in embedding physical domain knowledge, this invention constructs standardized multi-dimensional time series data, performs heterogeneous feature grouping and parallel encoding to obtain multiple sets of feature representations, then extracts and integrates features from multiple sources to obtain intermediate state features, and finally performs multi-branch prediction and adaptive weighted fusion based on the intermediate state features to obtain the final time series forecast result. This invention can achieve active focusing on abrupt events in industrial sequences and accurate characterization of extreme value amplitudes, significantly improving prediction accuracy, physical consistency, and robustness to extreme conditions.

The application discloses an emotion recognition method based on online cross-modal knowledge distillation, comprising the following steps: acquiring electroencephalogram and electrocardiogram original signals and windowing and cutting; constructing electroencephalogram and electrocardiogram student models, extracting intermediate features from each modal data segment through an encoder, and obtaining non-normalized prediction output through a classifier; constructing a teacher probability distribution through a joint encoder fusion; introducing adaptive contrast loss to align the cross-modal intermediate features, introducing distillation loss to constrain the prediction probability distribution of each modal to align with the teacher probability distribution; synchronously optimizing new student model parameters through online collaborative training; and performing actual inference prediction based on the student model after training. The application combines double modal signals to make up for the defects of single modal information, excavates the complementarity of modes, realizes dynamic generation of teacher supervision signals and real-time collaborative learning of modes through online distillation, does not increase test calculation overhead, effectively improves the recognition accuracy, model robustness and generalization ability, and has good application prospect.

The invention relates to the technical field of virtual power plant operation and optimization control, in particular to a virtual power plant-oriented multi-data-source fusion multi-energy load prediction method, which comprises the following steps of: constructing aggregated multi-source heterogeneous data of a virtual power plant; the data comprises historical multi-energy load data, calendar and market information and external environment auxiliary data in the jurisdiction of a virtual power plant, preprocessing the multi-source heterogeneous data, performing feature screening on the time sequence feature matrix based on correlation analysis, constructing a model input set, constructing a load prediction model, and predicting the load of the power plant. The model comprises an input layer, a plurality of stacked space-time fusion layers and an output layer which are connected in sequence; the space-time fusion layer comprises a multi-scale time convolution module and a graph convolution module which are parallel, and is used for model training and parameter optimization, load prediction and result output; and combined prediction of the electric load, the thermal load and the cold load is realized under a unified multi-scale space-time diagram neural network framework.

The invention belongs to the field of network social group analysis, and provides a multi-expert collaborative network social group description method based on modal dynamic fusion. The method comprises the following steps: firstly, inputting multi-modal original data of a network social group, and respectively capturing multi-modal private feature representations through a modal representation adaptive extraction module; then, deep fusion of cross-modal information is realized by utilizing a bidirectional state space model through a modal complementary information fusion module, and unified group multi-modal fusion representation is generated; thirdly, a KAN multi-expert network architecture is introduced through a multi-expert collaborative prediction module, different experts are adapted to diversified group characteristic modes such as mainstream, small crowds and temporary groups, and dynamic weights are calculated through a gating network to achieve accurate fusion of expert output; and finally, in combination with long-tail boundary perception loss function optimization model training, outputting a multi-dimensional feature description result of the group. According to the method, more comprehensive and accurate group feature representation can be obtained, and the depicting performance of the network social group in a complex scene is improved.

The application relates to the technical field of battery health management, and discloses a lithium battery SOH prediction method and system based on a Koopman gated residual neural network, which comprises the following steps: acquiring electrochemical impedance spectrum data of a lithium battery; based on an equivalent circuit model, first physical characteristics are fitted and extracted from the electrochemical impedance spectrum data, and a characteristic frequency point is determined; an electrochemical impedance imaginary part value at the characteristic frequency point is extracted as a second frequency domain characteristic; based on the first physical characteristics, an initial prior estimation value of a battery health state is generated, and the initial prior estimation value is dynamically trend-corrected to obtain a corrected prior estimation value; the first physical characteristics, the second frequency domain characteristic and the corrected prior estimation value are input into a pre-trained Koopman gated residual neural network model, and an adaptive gated fusion mechanism of the model is used to output a final prediction value of the battery health state. The application realizes relatively accurate and high-robustness online prediction of the battery health state.

This invention provides an artificial intelligence-based internet property rights transaction risk assessment and decision-making system, relating to the field of data processing technology. The system includes: a data acquisition and preprocessing module for real-time acquisition of raw transaction data, followed by cleaning and standardization; a multi-dimensional feature construction module for modeling by time windows and constructing a multi-level interaction relationship graph to obtain time-series feature data and graph-related feature data; and for normalizing and weighting the features; a risk identification module for extracting suspicious transfer behavior chains and tracing the flow of benefits, performing composite risk discrimination; and an intelligent decision-making module for performing hierarchical adaptive discrimination and case migration comparison, summarizing and generating risk assessment results and pushing early warnings; and receiving feedback from the management terminal, which is then used to optimize identification and decision-making parameters. This invention improves the autonomy and accuracy of property rights transaction risk assessment and decision-making.

The invention discloses an electrochemical device residual life prediction method based on a fusion model. The method comprises the following steps: S1, acquiring historical operation data of an electrochemical device, performing correlation analysis on the historical operation data, and screening out parameters strongly related to life attenuation as performance degradation indexes; s2, preprocessing the data of the performance degradation index; s3, predicting a degradation track of the preprocessed performance degradation index through a data-driven prediction model, and dynamically correcting the degradation track by fusing a physical experience correction model to obtain a degradation track correction fusion model as shown in a formula (1-1); and S4, determining the residual service life of the electrochemical device based on the corrected degradation track and a preset failure threshold value. According to the method, signal processing, machine learning optimization and an empirical model are fused, so that the residual service life of the electrochemical device is quickly and accurately predicted.

This invention discloses a video summarization method based on two-layer routed sparse attention and spatial pixel recalibration, belonging to the field of computer vision. The method includes: reading the input video and extracting frame-level feature vectors; constructing a video summarization generation model, performing channel enhancement on the frame-level feature vectors to obtain enhanced features; performing two-layer routed sparse attention and spatial pixel recalibration on the enhanced features, and fusing them to obtain fused features; inputting the fused features into an importance scoring regression network, outputting frame importance scores, and selecting frames to generate video summaries. This invention effectively fuses global dependencies and local details by combining two-layer routed sparse attention and spatial pixel recalibration, accurately identifying key segments in the video while reducing computational complexity. Experimental results based on the SumMe and TVSum benchmark datasets demonstrate that the method of this invention exhibits good performance in video summarization tasks.

The invention discloses an emotion recognition method based on online cross-modal knowledge distillation. The emotion recognition method comprises the steps that electroencephalogram and electrocardio original signals are obtained and subjected to window segmentation; constructing an electroencephalogram and electrocardio student model, extracting intermediate features from each modal data segment through an encoder, and obtaining non-normalized prediction output through a classifier; teacher probability distribution is constructed through joint encoder fusion; self-adaptive comparison loss alignment cross-modal intermediate features are introduced, and distillation loss is introduced to constrain prediction probability distribution of each modal to align to teacher probability distribution; new student model parameters are optimized synchronously through online cooperative training; and performing actual reasoning prediction based on a student model after training. According to the method, bimodal signals are combined to make up single-modal information defects, modal complementarity is mined, teacher supervision signal dynamic generation and modal real-time collaborative learning are realized through online distillation, test calculation overhead is not increased, identification precision, model robustness and generalization ability are effectively improved, and the application prospect is good.

The invention relates to the technical field of model training optimization, and discloses a new energy truck carbon emission model optimization method and system based on a neural network, and the method comprises the steps: collecting the carbon emission influence characteristics of a new energy truck and the historical record of the carbon emission; performing response trend analysis on each carbon emission influence characteristic, and identifying reversible influence characteristics in the carbon emission influence characteristics; constructing a carbon emission prediction model and defining structural parameters; searching a group of reference structure parameters for the carbon emission prediction model based on an EPO algorithm; and carrying out iterative training and verification tuning on a carbon emission prediction model based on the reference structure parameters. The new energy truck carbon emission prediction model is improved in the aspects of input feature compression, structural parameter optimization and abnormal state self-adaption, and the prediction accuracy of carbon emission is enhanced.

The present application provides an asphalt pavement apparent disease evolution deduction method, system, device and medium, which belongs to the field of road engineering, transportation infrastructure operation and maintenance and intelligent detection technology, and comprises the following steps: constructing a unified space-time reference system, aligning and normalizing multi-source data in space-time; forming section-level disease quantification characteristics based on the recognition results of the inspection images and the structural indexes; event coding the maintenance measures of the road surface and pre-processing the section-level disease quantification characteristics corresponding to the time of the maintenance measures; constructing the correlation, time lag effect and spatial consistency between the disease types, disease characteristics and structural indexes to generate the collaborative evolution characteristics of the coupling relationship between diseases; constructing a deep time series prediction model to obtain the disease development trend at a specified time scale in the future, and forming the section-level disease evolution prediction result. The present application can depict the disease development law at different road sections and different time scales, and predict the future disease expansion.

The application provides a kind of source load power prediction method, device and electronic equipment, the method comprises: obtaining source load power data and multiple covariant data, source load power data is fused with multiple covariant data, and the multiple source time series data set after fusion is obtained;Determine the optimal lag time that each meteorological factor in the multiple source time series data set after fusion influences source load power, and determine the input time window length according to the optimal lag time;According to the input time window length, input sequence is extracted from the multiple source time series data set after fusion, and the input sequence is input into source load power prediction model to obtain the source load power prediction result of future time period.The application can solve the problem that the existing source load power prediction scheme is insufficient in describing the space-time coupling relationship between source load power and multiple covariants, and it is difficult to accurately extract key information, resulting in poor prediction stability.The application directly outputs the source load power prediction result of future time period through source load power prediction model, without relying on error correction, thereby improving the stability and reliability of prediction.

This invention proposes a brain-inspired navigation method based on DSI decoupling representation and composite potential field path optimization. The method includes: acquiring a state sequence and calculating a successor representation matrix; obtaining a DSI representation from the successor representation matrix; generating a navigation path sequence based on the DSI representation; constructing a total potential function from the navigation path sequence; generating an optimal path sequence based on the total potential function and simultaneously introducing obstacle information; using the optimal state sequence extracted from the optimal path sequence as a supervision signal to update the DSI representation using gradient descent, and storing the obtained corrected DSI representation in a calibration experience buffer; repeatedly executing online navigation, path optimization, and calibration steps using the corrected DSI representation in the calibration experience buffer, and obtaining the optimal brain-inspired navigation sequence after a preset number of iterations. This invention applies composite potential energy optimization and obstacle constraints to the initial path, resulting in a geometrically more compact and safer path.

This invention provides a method and system for precise source tracing of nitrogen compound pollution sources based on multimodal data fusion and graph neural networks, belonging to the field of pollution source tracing technology. This invention acquires and quantifies the multimodal fingerprint features of each typical pollution source to form a source feature matrix, obtains multimodal spectral data from downstream monitoring points to obtain actual monitored downstream fingerprint vectors, and constructs a spatial topology map based on the spatial topology between monitoring points. The source feature matrix and the spatial topology map are input into a graph neural network to obtain contribution weight vectors, and the predicted downstream fingerprint vectors are reconstructed accordingly. A composite loss function including spectral reconstruction loss, physical constraint loss, and sparsity constraint is constructed, and convergent contribution weights satisfying non-negativity and sum-to-one constraints are obtained through iterative optimization. This invention can output a physically meaningful and interpretable quantitative contribution rate of pollution sources for precise source tracing analysis in nitrogen compound pollution scenarios.

The application provides a kind of based on feature enhancement and composite model's ultra-short-term wind power prediction method, belongs to ultra-short-term wind power prediction technical field, this method includes: by calculating the mutual information of multiple meteorological characteristics and wind power output and normalizing, screening to obtain key features, and constructing wind power prediction fitness function, by minimizing wind power prediction fitness function, obtain enhanced input features;By introducing one-dimensional convolutional neural network and bidirectional long short gate recurrent unit, construct the residual network of bidirectional long short gate recurrent unit and convolutional neural network fusion, and introduce space-time attention mechanism, construct to obtain composite residual network ultra-short-term wind power prediction model;And use multiple meteorological characteristics and enhanced input features to predict to obtain ultra-short-term wind power prediction result;The application solves the problem that the influence of meteorological multi-scale fluctuation on output cannot be accurately grasped, prediction is disconnected, disturbance is not considered and prediction accuracy continues to decline with step increase.

The application discloses a high-resistant starch rice screening and identification and content detection method and system, and relates to the technical field of rice screening. The high-resistant starch rice screening and identification and content detection system comprises a rice screening and identification module and a rice content detection module. The application realizes non-destructive, high-throughput and high-precision comprehensive evaluation of the resistant starch of rice by constructing a multi-modal fusion screening system of starch metabolism fluctuation frequency analysis, three-dimensional seed twin modeling, virtual enzyme molecule diffusion simulation and bionic digestion dynamics school, can complete large-scale seed screening under the premise of not significantly increasing sample loss, and improves the ability to describe the dynamic behavior of starch metabolism by introducing frequency domain features and spatial topology analysis, so that the screening result not only reflects the static content difference, but also can represent the dynamic law of the formation process.

The application discloses a power distribution network voltage prediction method based on a space-time graph neural network, comprising the following steps: abstracting a power distribution network into a dynamic graph structure; obtaining a historical operation data sequence; and inputting the historical operation data sequence into a pre-trained space-time graph neural network model for prediction. The model extracts spatial features through multi-relation graph convolution and electrical guide graph attention network, extracts time features through a recurrent network, fuses the two by using a space-time cross attention mechanism, and finally decodes and outputs future multi-step voltage prediction values and uncertainties. By explicitly modeling the space-time coupling characteristics of the power grid and embedding physical constraints, the application significantly improves the accuracy, robustness and interpretability of voltage prediction under high-proportion distributed energy access, and can provide proactive decision support for active voltage control.

The application relates to the technical field of aerospace and artificial intelligence, and provides a space target orbit intelligent prediction method fusing an orbit dynamics constraint, which comprises the following steps: preprocessing original orbit data of a space target to generate a complete time series data set with continuity and uniform time intervals; constructing a physical information neural network model which represents input and output in the form of time series data, and performing differential calculation on an output sequence by using an automatic differentiation mechanism; designing and introducing a loss function with an adaptive weighting mechanism to dynamically balance loss terms of network modules and physical modules in a training process; training the physical information neural network model; setting a data subset sampling rate, constructing a multi-scale data scene, and evaluating performance of the physical information neural network model under different data scale conditions. The application can improve modeling capability for a long-time orbit evolution process and realize dynamic adjustment of weights of various loss terms.

This invention relates to the field of medical information security technology, and discloses a blockchain-based method and system for auditing and detecting unauthorized access to medical data. The method includes: structurally modeling medical data access behavior to construct access behavior feature vectors; performing multi-source fusion risk assessment based on an access control policy model and a behavior learning model to obtain a comprehensive risk value; executing a graded processing strategy according to the risk level and generating corresponding audit logs; structurally organizing the audit logs and constructing a Merkle tree to generate data fingerprints; writing the data fingerprints and summary information into the main chain, storing the complete audit logs in a side chain or off-chain storage, and achieving consistency verification through Merkle proofs; using smart contracts to verify the audit data and automatically execute access control responses, while simultaneously feeding back the audit results for model updates. This invention enables real-time detection, graded processing, and trusted auditing of medical data access behavior.

The invention provides a high time frequency method and device for building power consumption data and electronic equipment, and relates to the technical field of power systems. The high time-frequency method for the building power consumption data comprises the following steps: determining a predicted fluctuation amplitude of the power consumption data of each time step in a second time-frequency power consumption data sequence of a target building according to an amplitude characteristic parameter corresponding to the power consumption data of each time step in a first time-frequency power consumption data sequence; according to the fluctuation characteristic parameter of the power consumption data of each time step in the first time-frequency power consumption data sequence, determining a zero mean value standardized value of the power consumption data of each time step in the second time-frequency power consumption data sequence; and according to the predicted fluctuation amplitude and the zero-mean standardized value of the power consumption data of each time step in the second time-frequency power consumption data sequence, determining a power consumption data predicted value of each time step in the second time-frequency power consumption data sequence. According to the invention, the high-fidelity high-time-frequency power consumption data for the building can be generated based on the original low-time-frequency power consumption data of the building.

The invention discloses an intelligent coffee machine operation state monitoring method and system based on reinforcement learning, and the method comprises the following steps: S1, collecting and preprocessing data, and forming an operation parameter sequence; s2, carrying out dimension reduction processing on the operation parameters, and constructing an operation state sequence; s3, performing multi-step prediction on the running state sequence through a DLinear model, and generating a state prediction sequence by adopting a linear predictor; s4, calculating an instant reward value of each regulation and control behavior according to the running state sequence and the state prediction sequence; s5, an A3C algorithm is adopted, and a regulation and control instruction is generated and executed according to the instant reward value; s6, establishing a state transition group and writing the state transition group into an empirical data set; and S7, updating the DLinear model and A3C algorithm parameters according to the empirical data set. According to the method, the Markov model, the principal component analysis, the DLinear model and the A3C algorithm are fused, and the method has the advantages of being high in adaptability, high in prediction precision and good in stability.

The invention discloses a single-cell drug response prediction method based on two-channel comparative learning, and relates to the field of artificial intelligence assisted drug research and development. The invention aims to solve the problem that the existing single-cell drug reaction prediction precision is low. The method comprises the following steps: acquiring an original single cell transcriptome counting matrix corresponding to a to-be-detected single cell, extracting a to-be-detected gene set with an expression counting value greater than 0 from the original single cell transcriptome counting matrix, acquiring a gene identity embedding vector and an expression intensity embedding vector, and acquiring a cellular-level semantic feature vector by utilizing the expression intensity embedding vector; obtaining a structural feature representation vector of a single cell layer by using the gene identity embedding vector; and acquiring a drug response result and a corresponding probability by using the cellular-level semantic feature vector and the structural feature representation vector of the single cell level. The method is used for obtaining the single cell response reaction under the given drug condition.

The invention discloses a network performance hierarchical evaluation method based on a multi-agent system, and the method comprises the following steps: carrying out the hierarchical division of a to-be-evaluated wireless communication network, and constructing a hierarchical evaluation system; collecting performance data of the corresponding wireless network object; constructing a hierarchical evaluation graph; performing input embedding processing on each evaluation node in the hierarchical evaluation graph; generating a cross-level semantic distance coding result and a conflict perception bias result; inputting the node representation set, the edge attribute coding result, the cross-level semantic distance coding result and the conflict perception bias result into an improved Grapporter model, and executing multi-head attention calculation; carrying out step-by-step convergence on the evaluation node update representation set by utilizing a hierarchical virtual agent node chain; and correcting the overall performance evaluation result of the network, and generating a resource regulation and control strategy or an operation parameter optimization strategy. According to the method, multi-agent layered evaluation is adopted, and accurate optimization of the wireless network performance is realized.

A spatiotemporal multimodal quality prediction system for bananas and its training method are disclosed, relating to the field of intelligent detection and preservation information technology for agricultural products. This system alleviates the problems of existing banana quality prediction technologies, such as difficulty in comprehensively reflecting the intrinsic mechanisms of banana quality evolution and the tendency for error accumulation during long-term prediction. The spatiotemporal multimodal quality prediction system for bananas includes an image data acquisition module for obtaining visual spatial features based on the time series of images of the banana to be detected; a physiological data acquisition module for obtaining physiological features based on physiological data; a multimodal data fusion module for fusing the visual spatial features and physiological features through multimodal data fusion to obtain global temporal features; and a multimodal quality prediction module for using the global temporal features to obtain a predicted banana image and corresponding predicted physiological data through multi-task collaborative decoding. This invention is applicable to fields such as multimodal data fusion and non-destructive testing of agricultural product quality.

The invention discloses a network traffic probe anomaly detection method based on deep learning. The method comprises the following steps: acquiring network traffic data and recording an acquisition timestamp; executing time alignment processing to generate a unified time index, and performing association processing to obtain a traffic feature sequence; executing numerical value standardization processing and constructing a continuous time flow sequence; inputting an improved liquid state time constant network, and generating a flow state sequence under constraint; calculating a traffic change feature sequence based on the continuous time traffic sequence; based on a time constant mapping function, mapping the traffic change feature sequence to generate a time constant parameter sequence, and performing constraint to obtain an updated traffic state sequence; constructing a state evolution reference trajectory, and calculating a state evolution deviation degree sequence; and generating an exception indication sequence, and outputting an exception detection result. According to the method, network flow anomaly detection is realized by adopting liquid time constant deep learning, and the method has the advantages of high time adaptability and stable detection.

To address the issue of user viewing behavior being easily influenced by confounding factors such as video duration in short video recommendation, leading to distorted interest modeling, this paper proposes a short video recommendation method based on causal inference and graph convolutional networks. First, a do-operator is introduced to explicitly intervene in users' historical behavior, weakening the confounding effect of duration bias on interest expression and obtaining more realistic preference signals. Second, positive and negative sample sets are constructed, and representation learning and aggregation are performed separately through differentiated graph convolutional networks to achieve multi-dimensional user preference modeling. Furthermore, graph propagation mechanisms and expert network design are combined to enhance the model's ability to characterize complex interest structures. Experimental results show that this method can effectively alleviate confounding bias and improve recommendation performance and model robustness.

The present application relates to the technical field of industrial big data processing, artificial intelligence and industrial process control, and particularly relates to a sintering state prediction method based on dynamic graph and topological perception hierarchical interaction, which comprises the following steps: S1, constructing an initial sample matrix based on original time series data of a sintering industrial site; S2, constructing a dynamic adjacency matrix in combination with a historical state feature matrix; S3, grouping the time feature and the dynamic graph adjacency matrix, and sequentially performing in-group convolution and inter-group interaction convolution to extract a current time granularity spatiotemporal representation; S4, predicting future parameters according to the spatiotemporal representation and a control instruction feature matrix. The method of the present application can respond to non-stationary working conditions such as raw material property changes and equipment state fluctuations in a timely manner, and effectively overcomes the problem of sudden drop in prediction performance of traditional static graph models when the working condition changes.

The application discloses a power grid fault prediction method and system based on multi-dimensional analysis, relates to the technical field of power system operation state monitoring, and comprises the following steps: collecting multi-dimensional data and organizing the same to form a standardized multi-dimensional data matrix; constructing a dynamic grid abstract model based on the standardized multi-dimensional data matrix; extracting topological spectrum features of the dynamic grid abstract model, inputting the topological spectrum features and the multi-dimensional data into a multi-section hierarchical hybrid space-time attention network for deep fusion to form high-dimensional features after fusion; adopting a rule deduction method to identify fault probabilities of nodes and lines, obtaining power grid risk levels according to topological correlations between nodes and risk accumulation rules; and mapping the fault probabilities and the power grid risk levels to a visual risk topological graph to generate automatic early warning information according to node and line risks. The application realizes intuitive presentation and early warning prompt of power grid fault risks and improves power grid operation monitoring and fault prediction capabilities.

The invention discloses a DEM-based tunnel slope rockfall process simulation method, and relates to the technical field of numerical calculation of tunnel engineering technology, a simple rock slope containing DFN and a joint model is built, and microscopic parameters of the joint model are calibrated by simulating sliding, rolling and collision motions of rockfall and combining a friction coefficient and recovery coefficient calculation formula; on the basis of a geological section CAD drawing, discrete elements and CAD software are utilized to construct a complex slope and dangerous rock mass discrete element model, and a local particle encryption algorithm is adopted to optimize contact between a slope surface and dangerous rock mass particles; slope node data are converted through MATLAB, and a DFN and joint coverage model of segmented assignment is generated; and weakening a dangerous rock mass joint parameter simulation stripping process, and recording characteristics such as a falling rock movement track and speed. According to the method, refined simulation of the complex slope form and rockfall movement is realized, the model authenticity and the calculation precision are improved, the calculation efficiency is considered, and a reliable basis is provided for slope protection design.

This invention provides a method and system for identifying the lubrication state of self-lubricating spherical friction pairs, belonging to the field of mechanical equipment condition monitoring. The method collects acoustic emission signals during the operation of the friction pair using an acoustic emission sensor. After normalization, overlapping sliding window sampling, and data augmentation preprocessing, the signals are input into a dual-stream time-frequency fusion deep learning model. The model extracts complementary features through parallel time-domain and frequency-domain feature extraction modules, performs spatiotemporal modeling using a bidirectional GRU and Transformer encoder layer, and optimizes the model by combining a focus loss function and the AdamW optimization algorithm. Finally, the lubrication state prediction result is output. This invention fully utilizes the time-frequency characteristics of acoustic emission signals, achieving high accuracy, strong anti-interference capability, and good real-time performance. It can effectively identify the lubrication break-in period, stable period, and rapid deterioration period, providing a guarantee for the reliable operation of self-lubricating spherical friction pairs.