292results about How to "Easy to dig" patented technology

The invention provides a PCBA surface defect detection method and system based on image recognition, and the method comprises the steps: collecting and constructing a multi-class PCBA overall defect image data set, and carrying out the preprocessing; constructing a pre-training neural network model, and classifying the preprocessed PCBA overall defect image data set; constructing a local target identification method based on a YOLOX network according to an overall defect classification result; and carrying out local defect detection on the PCBA surface based on a local target identification method. According to the method, the overall classification and local detection strategies are comprehensively applied to reduce the burden of the model, and the detection precision and speed are improved. And an attention mechanism is introduced to improve the performance and generalization ability of the model. Data enhancement, consistency processing, classification labeling and the like are carried out on an image data set, and the defect detection effect of the model is improved. And multiple optimization strategies such as structure improvement, loss function improvement and pruning lightweight are adopted for the network model, so that the image target detection performance is effectively improved.

The invention provides a tunnel section identification and deformation detection method based on laser scanning point cloud data, and the method comprises the steps: directly processing the point cloud data in a three-dimensional space through polar coordinate conversion and local calculation weighted average modes to extract a central axis, so as to completely reserve the three-dimensional information of the point cloud data, and more accurately reflect the real spatial form of a tunnel. The problem that geometric feature information of the top and the bottom of the tunnel is lost due to projection is avoided. Tunnel section interception and section point cloud data determination are carried out on the basis, global and local deformation discrimination is carried out in combination with adaptive shape fitting, tunnel section deformation is accurately monitored and discriminated, possible tiny deformation of a tunnel in a local area can be sensitively captured, and even under the condition that overall deformation is not obvious, the tunnel section deformation can be accurately monitored and discriminated. Potential safety hazards can be found in advance, all-around and precise monitoring and judgment of tunnel section deformation are achieved, and powerful technical support is provided for long-term safe operation of a tunnel.

The invention belongs to the field of computer systems based on calculation models, and discloses a neural network model lithium ion battery state prediction method based on convolution and bidirectional long and short time memory, comprising the following steps: S1, data acquisition and data preprocessing; the method comprises the following steps: S1, constructing battery discharge data under a plurality of temperatures and different driving cycles by adopting a sliding window technology, S3, constructing a neural network model, namely a CNN-BiLSTM model, of a convolution and bidirectional long-short term memory network, S4, carrying out model migration, and S5, carrying out model migration according to an obtained CNN-BiLSTM-TL network model of a target domain. The SOC and the SOE can be accurately estimated, the charge and discharge control strategy of the battery can be optimized, the service life of the battery is prolonged, the efficiency of the battery is improved, and meanwhile the method is a key for guaranteeing efficient and stable operation of a battery system.

The invention discloses a self-supervised learning and multi-modal biomedical signal processing method, system and terminal, and belongs to the technical field of biomedical engineering and information, and the method comprises the steps: employing an unsupervised learning auto-encoder to extract the feature representation of multi-modal data; and taking the feature representation of the source domain data as input, taking the feature representation of the target domain data as output, and training the large language model based on the Transform architecture based on maximum likelihood estimation. According to the method, relevance and common features among multi-modal data are learned through an auto-encoder, and feature representation extracted by the auto-encoder is applied to pre-training of a large language model, so that the large language model learns more abstract and advanced feature representation, and the processing capacity and generalization performance of the multi-modal data are improved; meanwhile, due to the fact that an unsupervised learning method is adopted, a large amount of unmarked data can be used for pre-training the large language model, and therefore data resources can be better utilized.

The invention provides a railway signal monitoring data generation method and system, and relates to the technical field of railways, and the method comprises the steps: carrying out the time-space correlation fusion processing of obtained carrier frequency characteristics, response coding information, the receiving time of transponder signal data, train operation speed data and train positioning data, and generating a monitoring data record under a unified coordinate system; adopting a rule-based data compression algorithm to adaptively select a compression rule and a compression rate to perform data compression on the monitoring data record; packaging the compressed monitoring data record into a monitoring data file in a standard format; and the data management server decompresses and decodes the monitoring data file, cleans, classifies and indexes the data in the monitoring data file by adopting a MapReduce parallel computing framework, stores the structured monitoring data into a distributed database, and performs fragmentation and redundant backup management on the database.

The invention relates to the technical field of college teaching, in particular to an intelligent multi-dimensional college teaching quality improvement and evaluation system, which comprises a data acquisition module, a data processing module, an intelligent analysis module, a strategy recommendation module and an effect evaluation module, and is characterized in that the data acquisition module is used for acquiring original data in a teaching process from multiple dimensions; the data processing module is used for processing the collected original data to form a standardized and structured data set; the intelligent analysis module is used for carrying out deep analysis on the processed data by applying an intelligent algorithm and carrying out multi-dimensional quantitative evaluation on the teaching quality; the strategy recommendation module is used for providing personalized teaching strategy recommendation for teachers based on the analysis and evaluation result in combination with a preset teaching quality improvement model; and the effect evaluation module is used for tracking and evaluating the teaching effect after the teaching strategy is implemented in real time. The system realizes comprehensive, objective and accurate evaluation and improvement of teaching quality of colleges and universities.

The invention provides a corn genotype and environment cross-modal feature fusion genome prediction method and model, a deep learning method is adopted to model a complex nonlinear transformation relationship between different omics data including a genome and an environment group, and a GECFGP model is constructed to fuse corn cross-genotype and environment interaction features. Time sequence characteristics and important growth periods related to phenotypes are well mined, and the function of predicting the phenotypes of the complex agronomic traits of the corn is achieved. The method shows high robustness in different experimental settings, has superiority in genome prediction of G * E interaction, explores phenotype-related important growth periods and performance of predicting phenotypes across different years, provides a new thought for a genome prediction analysis method, and has wide application prospects. Therefore, theoretical and application value reference is provided for adaptability analysis of corn in different environments and exploration of key growth periods, and development of the field of intelligent breeding is accelerated.

The invention discloses a data processing method based on combination of data driving and a continuous casting production process, and belongs to the technical field of industrial data management. The method comprises the following steps: extracting multi-source heterogeneous data in the continuous casting production process according to an IT class and an OT class, and mapping a physical model of a physical sensor in a digital space; the Internet of Things equipment accesses collected data to an industrial gateway, an industrial grid carries out analysis, format conversion and downsampling on the data and then reports the data to a data engine, and the data engine stores multi-source heterogeneous data into a business database of a corresponding object model in a message queue publishing / subscribing mode; performing real-time online data management including abnormal state detection, data completion and the like; the production process data and the production materials are aligned in time and space; and a metallurgical affair map and a digital casting blank expansion map are constructed. According to the invention, online data management is realized, the data reliability is improved, and a data management scheme is provided.

The invention relates to the technical field of sea surface temperature prediction, and discloses a sea surface temperature prediction method based on a spatio-temporal enhanced ConvGRU, which comprises the following steps: preprocessing a sea surface temperature data set obtained by ERA5 reanalysis data of an European mid-term weather forecast center, and dividing the sea surface temperature data set into a training set, a verification set and a test set; inputting the data of the training set into a deep learning algorithm model of spatio-temporal position attention enhancement for training, using the data of the verification set to evaluate the model precision and adjust parameters, and using the test set to evaluate the generalization ability and prediction precision of the model; the model comprises a multi-scale feature extraction module, a position attention enhancement module and a convolution gating cycle unit module. And preprocessing sea surface temperature data of a sea area to be predicted, and inputting the sea surface temperature data into the qualified model for sea surface temperature prediction. According to the method disclosed by the invention, the processing capability of spatial information can be improved, and the prediction accuracy is improved.

The invention provides a software vulnerability type identification method based on long-tail incremental learning, and belongs to the technical field of software engineering. The technical problem that the vulnerability type recognition precision is reduced due to disastrous forgetting and long tail problems when an existing method is used for dealing with dynamic data streams is solved. According to the technical scheme, the method comprises the following steps: (1) searching a vulnerability type identification data set, and performing task division according to vulnerability publishing time; (2) selecting typical vulnerability samples from the old task data by using mixed playback; (3) designing a mixed prompt template and a mapping method, fusing source codes and vulnerability descriptions, and inputting the fused source codes and vulnerability descriptions into CodeT5 for model and prompt template training; (4) introducing an EWC regularization item; and (5) fusing the vulnerability code and description by means of the mixed prompt template to predict the vulnerability type. The method has the beneficial effect that the problems of disastrous forgetting and long tails are relieved.

The invention provides an unmanned aerial vehicle fault important characteristic parameter optimal selection method and system based on causal reasoning and a medium, and the method comprises the steps: S1, collecting fault data which comprises flight fault data collected by a gyroscope, a GPS, an airspeed meter and the like; s2, performing data processing on the fault data; s3, causal structure learning: performing causal structure learning on the data after data processing through a VAE decoder and a VAE encoder in sequence to obtain a weighted causal DAG representing a causal relationship between the features; and S4, important fault feature selection: after the weighted causal DAG is obtained, node scores are calculated, then the node scores are sorted in a descending order, the top F fault feature variables are selected as feature subsets, and F is a positive integer. The method can effectively screen out the characteristic parameters of the fault of the unmanned aerial vehicle, can be used for fault diagnosis, facilitates the reduction of the fault frequency of the unmanned aerial vehicle, and improves the safety and reliability of the unmanned aerial vehicle.

The invention discloses a wave prediction method based on MEEMD-ATTBiLSTM, and relates to the technical field of marine monitoring, and the method comprises the steps: carrying out the boundary extension of a training set through employing an LSTM algorithm, and recognizing adjacent first extreme points; extending the first extreme point into a symmetric closed sequence by using mirror image continuation, and forming a decomposition training set by using the symmetric closed sequence and the training set after boundary continuation; processing and decomposing the training set by adopting an empirical mode decomposition method to obtain a plurality of groups of imf components and a group of res residual data; frequencies of each group of imf components are different and are arranged from high to low; respectively inputting each group of imf component and res residual data into an Attention network, inputting the original sequence into the Attention network to establish a weight index, and obtaining a weight coefficient matrix of each group of imf component and res residual data relative to the training set; and inputting each group of weight coefficient matrixes into the Bi-LSTM model, predicting each group of imf components, and calculating and predicting the SWH according to the predicted imf components. According to the invention, short-term wave prediction can be realized, and the wave prediction effect is improved.

The invention relates to a plant non-histone lysine crotonylation site identification method and a plant non-histone lysine crotonylation site identification system. The method comprises the following steps: extracting non-histone lysine crotonylation Kcr sites from a protein sequence to form a data set; extracting a first feature vector by using a self-attention mechanism and a smoothing function through a protein language model ESM2; capturing a long-distance dependency relationship and context information in the protein sequence through a bidirectional long-short-term memory network, and extracting a second feature vector; and carrying out feature fusion processing and then carrying out classification prediction to obtain a Kcr site identification result. By constructing a protein sequence data set, ESM2 utilizes a self-attention mechanism and a smoothing function, and BiLSTM captures a long-distance dependency relationship and context information in the protein sequence, efficient feature extraction of the protein sequence is realized, and more comprehensive information can be provided through fusion processing, so that the detection precision of Kcr site detection and recognition is improved.

The invention relates to the technical field of multi-modal data fusion, and discloses a geochemical data prediction method and device based on multi-modal data fusion. The method comprises the following steps: inputting multi-modal geoscience data into a geochemical data prediction model to obtain geochemical prediction data; the multi-modal earth science data comprises vegetation information, a digital elevation model, aeromagnetic data and a remote sensing image; the spatial multi-modal extractor of the geochemical data prediction model comprises a plurality of feature extraction modules, and each feature extraction module is used for extracting spatial features of earth science data of one modal; the multi-head convolution attention layer is used for fusing the spatial features extracted by each feature extraction module; the spectral convolution module is used for extracting spectral features of the remote sensing image; and the feature fusion layer is used for fusing the multi-space features and the spectral features. According to the invention, accurate fusion of the multi-modal earth science data is realized, and the precision of data prediction is improved.

The invention relates to a protein structural domain ontology knowledge semantic annotation method and an online analysis engine system, and provides a protein structural domain semantic annotation map for a user by integrating protein structural domain data types with different definitions. According to the semantic annotation method, a mapping matrix between a structural domain / superstructural domain and an ontology term is constructed by using protein structural domain information and protein ontology annotation; performing overall and relative statistical inference based on the mapping matrix, and retaining the most relevant ontology terms; and obtaining the ontology annotation taking the structural domain as the center by following a real path rule, and for each structural domain / superstructural domain, screening an association relationship between terms according to the obtained overall and relative statistical inference results to form an ontology annotation configuration file. Compared with the prior art, the method has the advantages that a user is supported to carry out semantic annotation on the corresponding protein structural domain, meanwhile, enrichment analysis of ontology knowledge is supported, and powerful support is provided for exploration in the field of semantic annotation of the protein structural domain.

The invention relates to the field of conveying equipment, and discloses continuous conveying equipment for mine bulk material conveying and a control method thereof.The continuous conveying equipment comprises a dustproof shell, one end of the dustproof shell is fixedly connected with a first motor, an output shaft at one end of the first motor is sleeved with a conveying belt, and one end of the conveying belt is sleeved with a conveying belt set; a gear set sleeves one end of the conveying belt set, a rotating wheel is rotatably connected to the middle of the dustproof shell, a plurality of digging shovels are fixedly connected to the outer wall of the rotating wheel, the rotating wheel and the gear set are fixedly connected in a sleeving mode, a collecting box is fixedly connected to the top end of the dustproof shell, and a first guide plate is fixedly connected to the lower end of the collecting box; the bottom end of the collecting box is fixedly connected with a second guide plate, the bottom end of the second guide plate is located at the top end of the conveying belt, and the rotating wheel is in a circular truncated cone shape. Compared with the prior art, the digging shovel rotates to dig bulk materials, so that the bulk materials are continuously dug by the digging shovel, and the efficiency of digging the bulk materials is improved.

The invention discloses a user position completion method and device, terminal equipment and a storage medium. The method comprises the steps of obtaining user position track data; according to a pre-trained position supplement model, determining position trajectory data needing to be complemented in the user position trajectory data; wherein the pre-trained position supplement model is obtained by training the deep learning model by adopting original position trajectory data; the deep learning model is established based on a graph neural network and a bidirectional encoder, and the position supplement model is used for predicting and filling missing parts in the trajectory data; according to the position trajectory data needing to be complemented, the missing part in the user position trajectory data is complemented, accurate complementation of the user position trajectory data is achieved, the accuracy and integrity of user behavior prediction are improved, the complex behavior mode of the user in the space-time dimension can be better understood and mined, and the user experience is improved. And the capability of capturing the behavior characteristics of the user is enhanced.

The invention belongs to the technical field of punching equipment, and particularly relates to a special variable-diameter down-the-hole drill bit for large strip mine deep blast hole punching, which comprises a drill bit neck, a drill bit body is fixedly connected to the drill bit neck, first drill teeth are fixedly mounted on the drill bit body, square grooves are formed in the drill bit body in an array manner, and adjusting assemblies are arranged in the square grooves. A square block is arranged in the square groove in a penetrating mode, one end of the square block is fixedly connected with a reaming head, and a frame is arranged in the square block. According to the special variable-diameter down-the-hole drill bit for drilling the deep blast holes in the large strip mine, through the arranged square block and the reaming head, the blast holes with different diameters can be conveniently excavated for use, so that when the explosive consumption is increased and the blasting effect is ensured, the micro motor works to enable the square block to move, and through cooperation of the reaming head after the square block is moved, the drilling efficiency is improved; when the drill bit body rotates, the reaming head can be driven to do circular motion, the area of a digging and penetrating site of the drill bit body is enlarged through the reaming head, and the purpose of enlarging and adjusting the diameter of a dug blast hole is achieved.

The invention relates to an ardisia crenata carbon glycosyl transferase AcCGT64 gene and application of the ardisia crenata carbon glycosyl transferase AcCGT64 gene in preparation of bergenin, and belongs to the technical field of biology. The nucleotide sequence of the ardisia crenata carbon glycosyl transferase AcCGT64 gene is as shown in SEQ ID NO. 1, and the total length of the sequence is 1359 bp; the amino acid sequence of the encoded protein is as shown in SEQ ID NO.2, and 451 amino acid residues are encoded. The ardisia crenata carbon glycosyl transferase AcCGT64 gene disclosed by the invention can be used as a biosynthesis regulation gene of bergenin, is applied to preparation of bergenin, and is remarkable in application prospect and easy to popularize and apply.

The invention relates to a giantleaf ardisia rhizome oxygen methyltransferase AgOMT03 gene and application thereof in preparation of bergenin, and belongs to the technical field of biology. The nucleotide sequence of the giantleaf ardisia rhizome oxygen methyltransferase AgOMT03 gene is as shown in SEQ ID NO.1, and the total length of the sequence is 1095bp; the amino acid sequence of the encoded protein is as shown in SEQ ID NO.2, and 364 amino acid residues are encoded. The giantleaf ardisia rhizome oxygen methyltransferase AgOMT03 gene disclosed by the invention can be used as a biosynthesis regulation gene of bergenin, is applied to preparation of bergenin, and is remarkable in application prospect and easy to popularize and apply.

The invention discloses a deep hole blasting method for a strip mine step, relates to the technical field of blasting engineering, and aims to solve the problems that an existing strip mine step blasting method is easy to cause excessive large blocks on a slope surface and on the front section of the step, so that the excavation difficulty and the transportation potential safety hazard are increased. The method comprises the steps that an inclined first main blast hole and a vertical second main blast hole are drilled in the end, close to the slope surface, of a step to be blasted; a first auxiliary hole crossed with the first main blast hole is drilled in the slope surface, and a second auxiliary hole is formed beside the first main blast hole of the step; charging each blast hole with explosives; all the blast holes are connected into a detonating network through detonating lines; detonating through an exploder; wherein the first main blast hole and the second main blast hole can be distributed in a quincunx shape, the second main blast hole is vertical, and each main blast hole comprises a filling section and a charging section. According to the method, the auxiliary holes are formed to solve the problem of insufficient explosion stress of the slope and the step front section, and the explosion effect is improved.

The invention discloses a short-term power load prediction method based on image domain feature extraction, and the method comprises the steps: obtaining and preprocessing multi-source time sequence data of a power load, and forming a one-dimensional power load historical information sequence data set; converting a one-dimensional time sequence in the data set from a time domain to an image domain by adopting a GAF algorithm to form a two-dimensional power load historical information image data set; inputting images in the two-dimensional power load historical information image data set into a ConvLSTM (Convolutional Long-Short-term Memory) model to extract daily and weekly period and time sequence characteristics of the power load and carry out short-term power load prediction; meanwhile, an extrusion-excitation (SE) channel attention mechanism is introduced into the ConvLSTM model so as to consider the nonlinear influence of the characteristic variable on the power load change. According to the method, the prior knowledge and the similar daily method are utilized to predict the logic of the power load, the power load prediction precision is improved, and automatic intelligent auxiliary reference is provided for power system dispatching decision making.

The invention discloses an enterprise information sharing management method, system and device and a storage medium, which are convenient for managing enterprise data, can promote cooperation and communication among departments in an enterprise, and can improve the utilization efficiency and value of data. The method comprises the following steps: acquiring department data uploaded by each department, and judging whether the format of the department data uploaded by each department accords with an uploading format or not through a regular expression; when it is determined that the format of the department data accords with an uploading format, a unique identification identifier is automatically generated according to the department data, and the identification identifier comprises a timestamp and department information corresponding to the department data; classifying the department data according to department information to obtain classified data; uploading the classified data to an information management system; and when application information of target data is received, judging whether the target data has the access permission or not according to the application information.

The invention provides an economic prediction method and system based on a principal component regression model. According to the method, the advantages of principal component analysis and multiple regression are fused, and the defects existing in statistical method prediction are overcome by using the principal component analysis method, so that influence factors of economic development are better mined, and economic data are accurately predicted.

The utility model discloses a novel carrot seeding machine, which belongs to the technical field of carrot seeding and is characterized in that symmetrical rollers are arranged at the bottom of a seeding machine frame, a lifting plate is slidably connected to the inner wall of the seeding machine frame, an adjusting rod is rotatably connected to one surface of the lifting plate through a bearing seat, and symmetrical connecting rods are fixedly connected to the other surface of the lifting plate; according to the novel carrot seeding machine, through the arrangement of the motor, the transmission, the rotating shaft and the spiral blades, the discharge port can be prevented from being blocked during seeding, the rotating speed of the rotating shaft can be adjusted according to needs, then the discharge amount of seeds is accurately adjusted, the use convenience is greatly improved, and through the arrangement of an adjusting rod, a lifting plate and an opening, the practicability is high. The lifting height of the soil turning wheel can be adjusted according to needs, soil turning at different depth positions is facilitated, and in the carrying and moving process, the soil turning wheel can be prevented from being damaged due to contact with stone or a cement pavement on the ground.

The invention relates to a federal deep clustering method and system based on differential privacy, processing equipment and a storage medium, and is characterized in that the method comprises the steps that a central server trains a global model based on a public data set and a local clustering model of each client, and a trained global model is obtained; and each client adopts the trained global model to predict the local data set to obtain a corresponding clustering distribution result, so that the method can effectively relieve adverse effects caused by isomerism, improves the clustering performance of the model, and can be widely applied to the fields of data analysis and network security.

The invention belongs to the technical field of switch cabinet monitoring, and particularly relates to a high-voltage switch cabinet fault early warning method and system, computer equipment or a storage medium. The early warning method comprises the following steps: acquiring an ultrasonic signal, a transient ground voltage signal, an ultrahigh frequency detection signal, ozone concentration, oxygen concentration, an SF6 signal, a temperature signal and a humidity signal; performing data preprocessing on the acquired signal to obtain a processed signal; and inputting the ultrasonic signal and the preprocessed signal into the BP neural network model which is qualified in training, and outputting a prediction result. The operation state of the high-voltage switch cabinet can be monitored from multiple angles, possible limitation of single parameter detection is avoided, and the BP neural network has strong nonlinear mapping capability and self-learning capability, so that whether the high-voltage switch cabinet has a fault or not and possible types of the fault and the like can be accurately predicted; and potential fault hidden dangers can be found in advance.

The invention belongs to the technical field of plant cultivation, and particularly relates to a screening and fermentation method of a pepper root rot antagonistic bacterium W-5, which comprises the following steps: 1, preparing experimental materials, namely preparing a test strain, a culture medium, a reagent and an instrument; 2, separating, screening and identifying the antagonistic bacteria; 3, observing the influence of the antagonistic bacteria W-5 on the growth of the mycelia at the front end of the fusarium solani; step 4, detection of pathogenic fungus bacteriostasis spectrum by antagonistic bacteria W-5, the structure is reasonable, the main pathogenic fungus fusarium solani causing the pepper root rot disease is taken as an indicator bacterium, the pepper root rot antagonistic bacteria are obtained through separation by adopting a soil plate dilution method and a plate confrontation method, and the antagonistic bacteria are identified by combining morphology with molecular biology; meanwhile, the broad-spectrum antibacterial activity of the strain W-5 is measured, and the optimal fermentation condition of the strain W-5 is optimized based on a response surface method designed by combining a single-factor test with Box-Behnken, so that ginseng is expected to be provided for development and application of a pepper root rot antagonistic biological agent.

The invention relates to the technical field of information mining analysis, in particular to an information mining analysis method in an industrial scene, which comprises the following steps of: acquiring each data source, determining a data state according to a data source correlation degree and a mining difficulty coefficient, and determining a strategy analysis mode according to the data state; when the sub-data is extracted and analyzed, a processing mode is determined according to a dangerous sub-data distribution coefficient and a dangerous turbulence value to obtain a sub-data set, and the processing mode is a first extraction mode determined according to a dangerous end radiation coefficient and an in-end sub-data correlation coefficient, or a second extraction mode determined according to a dangerous gathering area category; when the associated source group is selected and analyzed, a clustering mode is determined according to the feature conversion coefficient and the number of the data sources, the data sources in the associated source group are selected according to the data flow anomaly index, and sub-data corresponding to the selected data sources serve as a sub-data set; the data mining efficiency in the industrial production process can be improved.