78 results about "Characteristic matrix" patented technology

The invention discloses a mucky soil foundation intelligent monitoring method based on multi-source data fusion, and particularly relates to the technical field of foundation monitoring and early warning. The method comprises the following steps: constructing an original soil body state data set of a mucky soil foundation, and carrying out anomaly elimination and time calibration based on a rheological constitutive model to generate a foundation state characteristic matrix; performing dynamic causal association tracking analysis on the feature matrix in a sliding window mode to generate a water-soil coupling drift parameter set; extracting a stress historical dependency coefficient set by using a nonlinear scale analysis method; and according to the water-soil coupling drift parameter set and the stress historical dependency coefficient set, the stability risk level of the mucky soil foundation is judged, and corresponding early warning information is selected from a preset graded early warning instruction database and output, so that the accuracy and adaptability of anomaly recognition and risk early warning are improved, and the accuracy and the adaptability of abnormal recognition and risk early warning are improved. The method is suitable for long-period stability monitoring and intelligent risk prevention and control of high-soft foundations or silt stratums.

The invention provides a feed mixing process uniformity dynamic evaluation and parameter adjustment method, and relates to the technical field of data processing, and the method comprises the steps: building a material characteristic matrix according to raw material characteristic data; correspondingly storing the material characteristic matrix and initial operation parameters of the mixing equipment to generate first mixing control data; in the mixing process, collecting local vibration amplitudes and temperature rise rates at different spatial positions, and combining the local vibration amplitudes and the temperature rise rates with the first mixing control data; calculating the vibration amplitude distribution difference in adjacent sampling periods in the feed mixing treatment process to obtain a spatial fluctuation index; when the spatial fluctuation index is continuously reduced and the temperature rise rate is periodically and suddenly increased, generating local accumulation indication data, and triggering short-time acceleration and inversion operation; executing feed mixing treatment, and generating a mixing termination signal when the spatial fluctuation index change rate of three continuous sampling periods is lower than a preset fluctuation threshold value; according to the invention, autonomy and accuracy of feed mixing uniformity adjustment are improved.

The invention provides a time-space evolution data processing method and system applied to tunnel deformation analysis, and the method comprises the steps: obtaining a time sequence monitoring sequence of different monitoring dimensions in a tunnel construction process, constructing a time-space evolution correlation model of tunnel deformation based on the time sequence monitoring sequence, and analyzing a dynamic coupling relation between structure response data and geological environment action data, generating a space-time correlation feature matrix, generating a feature evolution sequence of tunnel deformation based on the space-time correlation feature matrix, and performing nonlinear mutation feature extraction on the feature evolution sequence to obtain a mutation feature index set; according to the sudden change characteristic index set and a preset accumulated deformation rate threshold value, performing dynamic division of tunnel deformation safety levels, generating a safety level division result, and based on the safety level division result and the spatial distribution data of the tunnel construction section, executing spatial positioning and boundary contour extraction of the potential risk damage area. And obtaining a risk damage area distribution map. According to the invention, the tunnel construction safety risk can be accurately evaluated and effectively controlled.

The invention discloses a multi-layer heterogeneous sequential network feature alignment and clustering method and device based on tensor self-representation. The method comprises the steps that sequential feature matrix sequences {},..., {} of data views in a multi-view heterogeneous network in time windows are obtained; based on the matrix sequence of the first time window, constructing a Laplacian regular term by introducing a self-representation learning mechanism, modeling multiple views by adopting a tensor structure and introducing a Schatten p-norm regularization mode to construct an optimization objective function for optimization, and representing shared feature representation of each data view in the first time window; and solving the function and carrying out clustering processing on the obtained function, wherein the clustering result of each time window is used for carrying out characteristic analysis on nodes in the network. According to the method, on the premise that the original structure and semantics of the network are guaranteed, unified modeling can be carried out on the incomplete multi-view heterogeneous network, and feature expression is aligned.

The invention provides a penaeus vannamei quality prediction and grade evaluation method and system, and belongs to the field of penaeus vannamei cold-chain transportation process quality prediction, and the method comprises the steps: S1, constructing a penaeus vannamei data set, and carrying out the dynamic time-varying coupling correlation graph generation, convection-diffusion cooperative information propagation, multi-task joint optimization and end-to-end MLP prediction to obtain a penaeus vannamei data set; obtaining a prediction result of the quality of the penaeus vannamei; s2, splicing is performed according to the prediction result and the actual quality data, and a feature matrix is constructed; clustering the feature matrix by using a rule fusion K-means + + clustering method to obtain an optimal clustering center set; and S3, based on the prediction result and the clustering center set, carrying out dynamic prediction and evaluation on the quality of the penaeus vannamei in the refrigeration state in a period of time in the future. According to the invention, an intelligent solution is provided for quality monitoring in cold-chain logistics, and economic loss caused by quality deterioration is significantly reduced.

A generator fault identification method, device, product and medium relate to the technical field of generator fault category identification. The method comprises the following steps: acquiring a vibration signal and a noise signal; performing multi-mode decomposition processing on the vibration signal and the noise signal to obtain a first sub-mode component and a second sub-mode component; determining a vibration signal sequence based on each first sub-mode component and the vibration signal; determining a noise signal sequence based on each second sub-mode component and the noise signal; determining a vibration characteristic matrix based on the vibration signal sequence; determining a noise feature matrix based on the noise signal sequence; determining a first recognition result and a first confidence coefficient based on the vibration feature matrix; determining a second recognition result and a second confidence coefficient based on the noise feature matrix; when the identification results are inconsistent, determining a fault identification result according to the confidence coefficient; and when the identification results are consistent, determining the first identification result as a fault identification result. The method is advantaged in that generator fault identification accuracy is improved.

The invention discloses a photovoltaic array multi-working-condition mode recognition method based on Stacking and interpretability analysis. The method comprises the following steps: constructing a multi-dimensional photovoltaic array operation characteristic matrix; constructing a multi-dimensional photovoltaic array operation characteristic matrix and carrying out normalization processing; performing data set division on the normalized feature matrix and the corresponding tag set; the method comprises the following steps: constructing a Stacking learning framework based on a K-fold cross validation mechanism, selecting LightGBM, RF and XGBoost as base learners and LR as a meta learner, and constructing a multi-working-condition mode recognition model; and on the basis of the constructed multi-working-condition mode recognition model, taking F1 score maximization as an optimization target, and adopting an Optuna Bayesian search algorithm to carry out joint optimization on key hyper-parameters and fusion layer weights of all base learners. According to the method, deep derivation is carried out on the original monitoring quantity on the feature engineering level, a tree model-based Stacking architecture is constructed, and global and local interpretation mechanisms are introduced, so that the influence of environmental fluctuation and sample imbalance on an identification result is weakened, and the accuracy and reliability of photovoltaic array working condition identification are improved.

The invention discloses a geological exploration data analysis system and method based on multi-source data fusion, and relates to the technical field of data analys.The geological exploration data analysis method comprises the steps that a multi-source sensor is used for collecting general characteristic data in geology, and real-time characteristic values are calculated to construct an environment characteristic matrix; calculating the similarity between the real-time environment characteristic matrix and each type of geological data in the knowledge base, and selecting and judging to obtain a real-time survey geological type; respectively calculating data fusion weights of the real-time survey geological types by using the quality score and the relevancy; performing weighted fusion on the real-time data flow during geological survey by using the calculated data fusion weight; according to different types of geological historical survey data, abnormal thresholds of different types of geology are calculated and standardized, and an abnormal threshold database is generated; and extracting an abnormal threshold of the type of the real-time surveyed geology from the abnormal threshold database, judging the fused data by using the abnormal threshold, obtaining whether the real-time surveyed geology is abnormal or not, and performing early warning.

The invention provides an oil and gas geology virtual well construction method based on multi-modal rock debris digitization, which comprises the following steps: performing multi-dimensional physical and chemical characteristic detection on a rock debris sample of a target well to obtain characteristic parameters of internal composition and structure of the rock debris sample; standardizing the characteristic parameters to generate a digital characteristic matrix; inputting the digital characteristic matrix to a rock debris characteristic-logging parameter conversion model, and calculating a conversion relation between rock debris characteristics and logging data to obtain a pseudo-measurement curve; and acquiring logging data of an adjacent well, fusing the logging data and the pseudo-measurement curve by adopting a Kalman filtering method, and performing error correction on the pseudo-measurement curve to generate a virtual well. According to the invention, through multi-modal rock debris data acquisition and processing, in combination with the machine learning model and the Kalman filtering algorithm, accurate construction from the rock debris sample to the virtual well is realized, and the defect of insufficient construction precision of the virtual well is overcome.

The invention discloses an electrical load anomaly detection method and system based on a multi-granularity fuzzy rough set, and relates to the technical field of power data analysis, and the method comprises the steps: obtaining a high-dimensional time sequence feature matrix from to-be-detected electrical load time sequence data through employing a long-short term memory network; gathering the high-dimensional time sequence characteristic matrix into a plurality of multi-granularity pellets through a multi-granularity pellet generation method; calculating a multi-granularity fuzzy relationship among the multi-granularity pellets, constructing a multi-granularity fuzzy rough set model, and calculating the fuzzy rough density of each multi-granularity pellet and the multi-granularity fuzzy entropy of each attribute based on the model; calculating an abnormal score of each multi-granularity particle ball, mapping the abnormal scores of the multi-granularity particle balls to corresponding samples in the electrical load time sequence data, and performing abnormal load judgment on the corresponding samples based on the abnormal scores; the multi-granularity information of the load data can be effectively utilized, the anti-noise capability is enhanced, and the accuracy of anomaly detection is improved by processing the uncertainty of the data.

The invention provides a dissolved organic matter pedigree network construction method and system, and the method comprises the steps: obtaining a feature matrix and a molecular formula set of a dissolved organic matter sample; clustering the various samples according to the feature matrix to obtain a plurality of initial clusters; according to the feature matrix and the molecular formula set, determining genetic scores of various samples, including homogeneous family scores and heterogeneous family scores, and determining re-classified candidate samples according to the genetic scores; re-classifying the candidate samples, and iteratively confirming the re-classification of new candidate samples to obtain a plurality of target clusters; and according to the element change path between the internal samples of each target cluster, constructing a pedigree network of the dissolved organic matters. According to the method, pedigree affiliation identification and heterogeneous correlation analysis of the soluble organic matters are realized from two aspects of spatial distance distribution and molecular structure generality, and a pedigree network is constructed according to an element change rule, so that the reliability of the pedigree network is improved, and the resolution and stability of pedigree affiliation judgment are ensured.

The application provides a kind of metal pitting defect degree evaluation method and system based on artificial intelligence, by obtaining the micro porosity distribution data of metal plate in the forming process, and the internal fluctuation signal of metal plate under the action of directional stress wave;Based on the propagation attenuation rate of internal fluctuation signal, the geometric parameters of structure abnormal area corresponding to micro porosity distribution data are generated;Obtain the current density distribution data of structure abnormal area;Micro porosity distribution data, geometric parameters of structure abnormal area and current density distribution data are input into artificial intelligence model to associate micro porosity distribution data with current density distribution data, generate the pore current coupling characteristic matrix of metal plate;According to pore current coupling characteristic matrix, the quantitative evaluation result of internal pitting defect degree of metal plate is generated;The technical scheme provided by the application realizes the dynamic quantitative evaluation of the degree of metal internal pitting defect, and breaks through the technical bottleneck of traditional single index detection method in defect evolution mechanism analysis and quantitative precision.

The application relates to the technical field of titanium alloy detection, and discloses a titanium alloy wire performance detection method, equipment and system. The method collects real-time force value displacement data of a tensile test and echo signal data of ultrasonic detection, obtains smooth force value displacement data and frequency spectrum characteristic parameters after processing, fuses yield strength and tensile strength with peak frequency and amplitude attenuation rate in a characteristic layer, and constructs a first multi-dimensional characteristic matrix; the method inputs a pre-trained deep confidence network model, predicts mechanical properties such as elongation and hardness value, compares with a preset threshold value, marks unqualified items, and generates a detection report. The method improves the comprehensiveness and accuracy of detection through multi-source data fusion and advanced algorithm processing, realizes efficient evaluation of titanium alloy wire performance, and is suitable for strict detection requirements of titanium alloy wire quality in the fields of aerospace, medical devices and the like.

The invention discloses a defect identification method and system for a photoelectric detection image of power transmission and transformation equipment, and relates to the technical field of defect identification, and the method comprises the steps: collecting photoelectric detection data of target power transmission and transformation equipment; carrying out coupling analysis in combination with an equipment topology connection relationship, and drawing up a feature matrix; meanwhile, identifying a defect type mode based on a device topology connection relation and a change trend of a gray scale deviation coefficient in a plurality of detection periods; based on the characteristic matrix, combining the defect type mode and the photoelectric detection data to configure a defect confidence factor; and determining a defect severity level by using the defect confidence factor and carrying out graded reminding. The technical problems that in the prior art, electric transmission and transformation equipment photoelectric detection data integration is insufficient, the defect recognition precision is low, and operation and maintenance response lacks pertinence are solved, and the technical effects that integration analysis of multiple types of photoelectric detection data of the electric transmission and transformation equipment is achieved, and the defect recognition accuracy and the operation and maintenance response pertinence are improved are achieved.

The invention discloses a time-resolved metal material tensile strength detection device and method, and belongs to the technical field of metal material mechanical property detection. The method comprises the following steps: 1, configuring sample spectrum acquisition by using a time-resolved metal material tensile strength detection device; 2, respectively constructing a spectrum matrix Mn under different delay time; 3, sequentially carrying out saturation removal, abnormity removal, background removal and normalization on the spectrum matrix to form a sample initial feature vector; 4, respectively acquiring training set feature matrixes under different delay times, and forming training set feature input of the prediction model by adopting a variable screening and recombination mode; 5, performing multi-fold cross validation by using the randomly selected part of samples to adjust hyper-parameters so as to obtain a prediction model; the trained prediction model is used for metal material tensile strength detection; compared with the prior art, the technical problem of low accuracy in metal material tensile strength detection is solved by adopting the time-resolved spectrum.

The application discloses a liquid level nonlinear dynamic characteristic construction method, equipment and a storage medium, and relates to the technical field of system control. The method comprises the following steps: preprocessing input data and output data of a system at any moment under a steady state working condition to construct an input data state vector and an output data state vector; constructing a system time sequence basic structure function according to the input data state vector and the output data state vector; rewriting the system time sequence basic structure function into a system state space equation by constructing a state time sequence vector and a time factor decay amount; and solving a nonlinear dynamic characteristic matrix of the system based on the system state space equation. The application constructs a state vector and a time sequence basic structure function, reduces the dependence on a mechanism model, constructs a state space equation by using data, avoids a complicated model simplification process, directly extracts dynamic characteristics of the system from input and output data, and does not need to statistically obtain a large number of structural parameters.

The embodiment of the invention provides an interface element importance degree evaluation method and device based on fuzzy hierarchy analysis, and the method comprises the steps: collecting index data corresponding to interface elements at a current preset hierarchy through receiving an index structure corresponding to a fuzzy complementary judgment matrix, and constructing the fuzzy complementary judgment matrix based on the total element number and the index data; based on the total element number and matrix elements in a fuzzy complementary judgment matrix corresponding to the current preset level, determining factor weights of the interface elements in the fuzzy complementary judgment matrix, and based on the interface elements, determining a feature matrix corresponding to the fuzzy complementary judgment matrix; and determining the compatibility between the current fuzzy complementary judgment matrix and the feature matrix based on a compatibility calculation rule, and marking the current layout data as important layout data under the condition that the compatibility is not greater than a preset compatibility threshold value. According to the method, the defect of insufficient evaluation precision of the importance degree of the interface elements in the traditional technology is effectively overcome, and the accuracy of the importance degree of the interface elements is remarkably improved.

The invention discloses a rock permeability prediction method, a medium, equipment and a product, and relates to the technical field of permeability prediction.The method comprises the steps that multi-dimensional logging characteristics are obtained; constructing a Pearson's correlation coefficient matrix thermodynamic diagram to quantify the linear coupling strength of the multi-dimensional logging characteristics and the permeability, obtaining the correlation coefficient of the logging characteristics and the permeability of each dimension, screening the logging characteristics from the multi-dimensional logging characteristics according to the correlation coefficient, and constructing a characteristic matrix; building a parallel permeability model and a vertical permeability model based on a table prior data fitting network, training the parallel permeability model by using the feature matrix, splicing the parallel permeability predicted by the trained parallel permeability model as a new feature to the feature matrix, and training the vertical permeability model by using the spliced feature matrix; and predicting the rock permeability by using the trained parallel permeability model and vertical permeability model. The method can solve the problems that an existing method has high requirements for data quantity and quality and is weak in model generalization ability.

The invention discloses a bolt looseness detection method and device for a railway vehicle and a medium, and relates to the technical field of rail traffic operation and maintenance, and the method comprises the steps: carrying out the working condition coupling analysis of a time sequence image feature matrix, and generating a bolt state report through a lineation region feature fusion algorithm; extracting lineation fracture characteristics in the bolt state report, performing looseness risk grade division through multi-scale continuity analysis, and generating a bolt looseness risk distribution diagram; performing space coordinate mapping on the bolt loosening risk distribution diagram and the bolt area image data to generate a structured maintenance work order; and executing the structured maintenance work order, obtaining the looseness risk level after maintenance, and generating a maintenance verification report by comparing the bolt looseness risk distribution diagrams before and after maintenance. According to the method, through multi-frame time sequence image analysis and working condition coupling, complex environment interference is overcome, and the detection reliability is improved; based on a full-process predictive maintenance strategy, the detection efficiency is improved, and the maintenance cost is reduced.

The embodiment of the invention provides a counting method and device for composite actions of an excavator and electronic equipment. The method comprises the steps that a pilot pressure matrix of the composite action of the excavator is obtained; wherein the pilot pressure matrix represents pilot pressure values of actions in composite actions of the excavator; performing feature processing on the pilot pressure matrix so as to perform smoothing processing on the pilot pressure matrix to obtain a pressure feature matrix; wherein the pressure characteristic matrix represents the characteristics of the pilot pressure value of the composite action of the excavator; similarity calculation processing is conducted on the pressure characteristic matrix and a preset composite action matrix, and the number of composite actions of the excavator is obtained; wherein the preset composite action matrix is the characteristic of the standard pilot pressure value of the composite action. The method is used for achieving the effect of accurately and efficiently counting the composite actions of the excavator.

The invention relates to the technical field of electric digital data processing, and discloses a combined bridge interface damage time-varying analysis method and system, and the method comprises the steps: obtaining a slip sequence and temperature gradient data of a joint interface; extracting a high-frequency dynamic response signal through dynamic residual filtering; calculating a displacement compensation amount based on the thermal stress characteristic matrix, and eliminating the compensation amount from the dynamic response signal to determine a net slip characteristic value; mapping the slip space distribution difference into a topological feature vector by using an interface stress distribution mapping operator; analyzing the unsteady oscillation characteristics of the dynamic response signal, determining the cementation failure probability, and outputting a damage analysis index in combination with a topological feature vector, so that accurate decoupling of environment thermally induced displacement interference and a structure damage signal is realized, the influence of non-damage factors is eliminated, and a rigidity degradation critical point is captured by utilizing an interface topological evolution mechanism; the early interface void recognition reliability is improved, and bridge safety evaluation is supported.

The invention relates to the technical field of information security, in particular to a score query method and system based on biological characteristic matrix code identity verification, and the method comprises the steps: obtaining a score query request initiated by a user through a score query platform, and enabling the score query platform to call a matrix code management platform to generate and display a dynamic matrix code; scanning the dynamic matrix code through a matrix code verification APP, and collecting on-site biological characteristics of a user; comparing the field biological characteristics with a pre-stored biological characteristic matrix code file to generate a verification result; and if the verification result is successful, the matrix code verification APP sends an encryption confirmation message to the score query platform, and the score query platform calls and displays corresponding score data according to the encryption confirmation message. Through double-platform cooperation and a dynamic matrix code mechanism, high-security identity verification in a score query scene is realized, and identity false use, information leakage and unauthorized query are effectively prevented.

The invention discloses an SF6 high-voltage circuit breaker operation refusal risk prediction method, device, equipment, medium and product, and the method comprises the steps: carrying out the preprocessing of historical record data of an SF6 high-voltage circuit breaker in a target region, obtaining a plurality of spatial-temporal characteristic factors, mapping the spatial-temporal characteristic factors to a unified matrix space, constructing a spatial-temporal characteristic matrix, and carrying out the prediction of the operation refusal risk of the SF6 high-voltage circuit breaker. Constructing a key density function of each spatio-temporal characteristic factor based on the spatio-temporal characteristic matrix, and constructing a criticality fuzzy inference model based on the key density function, the criticality fuzzy inference model being used for quantifying fuzzy information of the spatio-temporal characteristic factors into a rejection risk degree, based on the criticality fuzzy reasoning model, the rejection risk prediction is carried out on the SF6 high-voltage circuit breaker in the target area, the high-precision prediction is carried out on the rejection risk distribution of the SF6 high-voltage circuit breaker from the space-time dimension, and the potential risk of the SF6 high-voltage circuit breaker is effectively captured in time, so that powerful support is provided for ensuring the stable operation of a power system.

The invention relates to the technical field of conductive material prediction, and discloses a conductive sponge conductivity accurate prediction method and a medium. According to the method, multi-dimensional feature data such as density distribution features, pore structure features and material component features of the conductive sponge are collected, then normalization processing is carried out on the data, dimensional differences are eliminated, and a standardized feature matrix is generated. Thirdly, the standardized feature matrix is input into a feature selection model, and key feature subsets are screened out by calculating feature importance scores; and constructing an initial conductivity prediction model by utilizing the key feature subset, and carrying out iterative optimization on the initial conductivity prediction model by adopting an integrated learning method so as to generate an optimized conductivity prediction model. And finally, inputting the multi-dimensional characteristic data of the conductive sponge, which is acquired in real time, into the optimized model, so that an accurate conductivity prediction value can be output. The method effectively improves the accuracy and reliability of conductivity prediction of the conductive sponge.

The present application belongs to the technical field of foundation pit engineering monitoring, and relates to a kind of foundation pit intelligent monitoring test method and equipment, which constructs the multi-physical quantity monitoring sequence of time sequence alignment by supporting member mechanical response parameter and medium mechanical response parameter, and establishes mechanical transmission path combined with spatial geometric relationship, can bring originally dispersed monitoring data into the same transmission link for analysis;By extracting the load transmission coefficient and constraint stiffness characteristics of each constraint node, a supporting member-medium coupling response characteristic matrix is constructed, providing a unified feature base for the establishment of theoretical transfer function, and based on the theoretical transfer function at each constraint node, the corresponding overall theoretical response envelope is established, which avoids false alarm or false alarm caused by relying on fixed threshold;By the over-limit combination state of supporting member response deviation and medium mechanical response deviation, the abnormal cause is determined as supporting member problem, medium change problem or supporting member and medium coupling failure problem.

The invention belongs to the technical field of steel structure floor design, and particularly relates to a transient response analysis method for converting structural damping into equivalent viscous damping in a full frequency band, which comprises the following steps of: 1, establishing a finite element model of a structure, and outputting characteristic matrixes of the structure, including a stiffness matrix and a mass matrix; 2, performing modal analysis based on the finite element model of the structure to obtain a modal matrix of the structure; 3, calculating a generalized stiffness matrix and a generalized mass matrix of the structure based on the stiffness matrix, the mass matrix and the modal matrix; step 4, calculating viscous damping of each order vibration mode of the structure based on the generalized stiffness matrix; 5, calculating the transient response of each order vibration mode of the structure based on the generalized stiffness matrix, the generalized mass matrix and the viscous damping of each order vibration mode of the structure; and 6, converting the transient response of each order vibration mode of the structure to a physical coordinate to obtain the transient response of the structure at each moment.

The invention discloses a transformer virtual diagnosis method, which comprises the following steps: acquiring monitoring data of a transformer at a plurality of continuous moments before the current moment, distributing the monitoring data to corresponding component nodes in an atlas model to obtain a characteristic matrix at each moment, inputting the characteristic matrix and the atlas model into a multilayer graph attention network to obtain a state embedding matrix at each moment, respectively inputting a GAT-VAE model to obtain an implicit state variable at each moment; deducing virtual monitoring data of the transformer at a plurality of continuous moments after the current moment, and inputting the virtual monitoring data and all implicit state variables into the time sequence prediction model to obtain an implicit state variable at each moment; calculating a virtual health index of each monitoring part at each moment after the current moment; and inputting the future implicit state variables and the virtualization diagnosis indexes corresponding to the future implicit state variables into the fault classification model to obtain the fault type probability of each monitoring component at each moment after the current moment, and determining the fault type of the monitoring component in combination with the virtual health indexes of the monitoring components at the corresponding moments.

The invention discloses a method for predicting coke quality of single coking coal of a test coke oven, which comprises the following steps of: S1, collecting historical test data of single coal, and constructing a characteristic matrix containing coal quality characteristic parameters and a label vector containing a coke quality index measured value; s2, analyzing data distribution characteristics of the measured values of the coke quality indexes, and determining an optimal clustering number and a classification boundary in combination with clustering validity evaluation; s3, according to the optimal clustering number and the classification boundary, category labels are distributed for historical experimental data, and a semi-supervised parallel classification model is trained by taking the feature matrix as input; s4, dividing the historical experimental data into a plurality of same proton sets according to the classification boundary, screening out key features, and constructing a cascade prediction model; and S5, extracting coal quality characteristic parameters of a to-be-predicted single coal sample, processing the coal quality characteristic parameters into feature vectors, inputting the feature vectors into the semi-supervised parallel classification model to determine the category of the feature vectors, calling the cascade prediction model corresponding to the category, and outputting a predicted value of the coke quality index.

The invention relates to the technical field of tea identification, and discloses a Pu'er raw tea year prediction method and device, electronic equipment and a storage medium, and the method comprises the following steps: obtaining spectral data and chemical component data of Pu'er raw tea; performing principal component analysis on the spectral data to obtain a spectral principal component characteristic matrix; performing mutual information maximization feature fusion on the spectrum principal component feature matrix and the chemical component data to obtain a fusion feature matrix; and based on the fusion feature matrix and a preset year prediction model, obtaining a prediction year of the Pu'er raw tea. According to the method, mutual information maximization feature fusion is performed on the spectral data and the chemical component data of the Pu'er raw tea, and the Pu'er raw tea year prediction is performed based on the year prediction model with high precision and strong interpretability, so that the prediction accuracy is remarkably improved.

The invention relates to the technical field of welding structure monitoring, and provides an intelligent ultrasonic monitoring method for creep damage of a dissimilar steel welding joint, and the method comprises the steps: calculating a waveform distortion index based on a time-frequency characteristic matrix; comparing the waveform distortion index with a self-adaptive threshold value in a monitoring time window, and identifying and marking potential damage signals; performing morphological analysis on the potential damage signal to obtain an envelope contour and a modulation mode, determining the existence of creep damage by adopting a pre-constructed damage verification model, and outputting a damage severity estimated value; integrating the spatial position and timestamp information of the potential damage signal, and executing a density clustering algorithm to detect a damage diffusion mode; and correcting the damage severity estimation value according to the diffusion mode attribute and reconfiguring the monitoring time window. Therefore, the problem that the creep accumulation condition and the residual life are difficult to accurately evaluate due to the fact that the early creep damage characteristics cannot be accurately identified due to the fact that a single method is adopted in ultrasonic data analysis in the prior art is solved.