54results about How to "Achieve migration" patented technology

The invention relates to the technical field of data processing, and discloses a multi-granularity extraction and enhancement method, device and equipment for multi-source heterogeneous data. The method comprises the following steps: performing multi-level feature extraction on a multi-modal data set and original feature parameters to obtain a multi-modal feature set; performing correlation analysis on the multi-modal feature set to obtain a feature correlation matrix and optimized feature representation; performing cross-modal knowledge migration processing on the feature incidence matrix and the optimized feature representation to obtain migration feature parameters; performing dynamic fusion processing on the migration feature parameters to obtain target fusion features; and performing multi-task segmentation processing according to the target fusion features to obtain a plurality of sub-task sequences, performing time sequence enhancement and multi-task fusion analysis on each sub-task sequence to obtain target extraction data, and performing visual display on the target extraction data. According to the method and the device, the efficiency and the accuracy of multi-granularity extraction and enhancement for multi-source heterogeneous data are improved.

According to the transfer learning joint inversion method and system provided by the invention, cleaning and gradient standardization processing are carried out on actual data, a stable high-quality actual data set is constructed, and for actual detection data without labels, the real detection data without labels can be subjected to gradient standardization processing; a cross-domain feature alignment method of a network end and an output end based on synthetic data and actual data is provided, a transfer learning effect between different data domains is tested by constructing a disturbance learning mode of matching different advanced prediction methods with training, and weak supervision transfer learning is performed on a small amount of manually labeled actual data. And parameter learning of the network in actual data is further optimized through model fine tuning.

According to the application of the bacteriophage to reduction of cholera transmission, the bacteriophage phi YSZKA is preserved in the China Center for Type Culture Collection on August 1, 2018, the preservation number is CCTCC NO: M 2018513, and the bacteriophage phi YSZKA is classified and named as klebsiella pneumoniae bacteriophage phi YSZKA. The method solves the problems of secondary pollution, wide transmission of vibrio cholerae and the like of the traditional split bacteriophage. The application of the bacteriophage can reduce the toxin production efficiency of vibrio cholerae in oceans, soil and sewers, thereby providing more possibilities for prevention and control of cholera epidemic situation and alleviation of cholera symptoms.

The invention provides a knowledge distillation passive domain unsupervised domain self-adaptive method, which realizes classification of a target domain without using source domain data by designing a teacher-student model and a competition mechanism. Firstly, a teacher network composed of a feature extractor, a classifier and a rotation classifier is constructed, source domain data is used for training, classifier parameters are fixed, a source model is finely adjusted through a self-supervised rotation sampling method, and a pseudo label of target domain data is generated through a clustering method. And then designing a student network composed of a feature extractor and a classifier, generating a pseudo tag only by using target domain data, and optimizing the discrimination and diversity of the tag by maximizing mutual information loss. And finally, screening the pseudo labels of the teacher-student network by using a competition module, and extracting the pseudo labels with higher confidence to update the student network, thereby completing knowledge migration. According to the method, source domain data is not needed, data storage and transmission cost is reduced, privacy and data security are improved, and unsupervised classification learning on unlabeled target domain data is supported.

The invention discloses a fault diagnosis method based on multi-dimensional knowledge migration. The method comprises the following steps: acquiring a source domain data set and a multi-target domain data set; constructing a double-branch neural network model based on guidance of data knowledge and semantic knowledge, training the double-branch neural network model, and optimizing model parameters by taking minimization of a total loss function of the double-branch neural network model as an optimization target; wherein the model comprises M + 1 data knowledge guidance branch networks and M + 1 semantic knowledge guidance branch networks; and performing fault diagnosis tasks of different target domains by using the trained double-branch neural network model. According to the method, the problem of negative migration caused by forced alignment distribution in traditional migration learning can be solved, and effective knowledge migration from a source domain to a multi-target domain is realized from the perspective of multi-dimensional knowledge fusion, interaction and driving; the method can be applied to industrial fault diagnosis tasks of industrial operation and maintenance and rapid migration deployment in various migration learning scenes, and has a good learning effect and generalization ability.

The invention discloses an image recognition method for confrontation domain self-adaption and a conjugate gradient kernel extreme learning machine. The method is characterized by comprising the following steps: firstly, constructing a feature extractor to obtain initial features of a source domain and a target domain; and then constructing an adversarial domain adaptive network to optimize two domain features, enabling a source domain feature extractor and a target domain feature extractor to play a game with a discriminator so as to reduce inter-domain differences between a target domain and a source domain, projecting the inter-domain differences to a common feature space, and finding an optimal target domain feature. And finally, constructing a classifier to recognize the image, and constructing a conjugate gradient kernel extreme learning machine as a classifier to classify the optimal target domain features to obtain a classification result. The invention aims to provide a lightweight network, namely an adversarial domain adaptive and conjugate gradient kernel extreme learning machine, which solves the problem of dependence of existing deep learning on a large amount of annotation data, reduces sample annotation cost in image recognition, reduces hardware implementation cost, realizes cross-domain knowledge migration, and completes unsupervised and high-precision image recognition of a target domain.

The invention discloses an inspection path determination method and device for electric power inspection and electronic equipment. The method comprises the steps that inspection task information of a plurality of inspection target objects to be subjected to electric power inspection is acquired, the inspection task information at least comprises position information and inspection duration information, and the position information is used for representing the positions of the inspection target objects; an inspection path is determined based on the inspection task information by using a pre-trained deep learning model, an electric power inspection path is obtained, the pre-trained deep learning model at least comprises a first network model and a second network model, the input end of the second network model is connected with the output end of the first network model, and the input end of the second network model is connected with the output end of the second network model; the parameter number of the first network model is greater than that of the second network model, and the inspection starting point and the inspection ending point in the power inspection path are at the same position. According to the invention, the technical problem of low efficiency of determining the electric power inspection path in the prior art is solved.

The invention discloses an eSIM migration method, device and equipment of a satellite network terminal, which are applied to an eSIM data migration scene between a first terminal and a second terminal, and the method comprises the following steps: in response to an eSIM data migration request initiated by a user, establishing a first communication connection channel through a seed number of an eSIM card of the first terminal; establishing a second communication connection channel through the service number of the eSIM card of the second terminal; the eSIM card of the second terminal is stopped based on the second communication connection channel, card data obtained through matching are written into the eSIM card of the first terminal through the first communication connection channel, and the card data correspond to the service number of the eSIM card of the second terminal. According to the eSIM data migration method and device, the communication connection channels between the different terminals and the over-the-air card writing subsystem are established respectively, eSIM data migration between the new terminal and the old terminal is achieved, eSIM card updating and management are facilitated, abnormal conditions occurring in the migration process are effectively handled, the migration safety and stability are improved, and the data migration requirement of a user is met.

The invention discloses a cross-regional trajectory prediction method and system based on class granularity space-time migration, and relates to the technical field of space-time data mining in an intelligent traffic system.The method comprises the following steps that traffic trajectories of vehicles are collected, the traffic trajectories are classified according to sampling timestamps, and source region data and target region data are obtained; wherein the source area data comprises source area historical data and source area future data, and the target area signal comprises target area historical data and target area future data; an initial trajectory prediction model is constructed, the initial trajectory prediction model is trained through the source area data and the target area data, a trajectory prediction model is obtained, and the initial trajectory prediction model comprises an embedding layer, a class granularity space-time module, a distribution converter and a reverse embedding layer; and collecting a real-time traffic track, and performing traffic track prediction by using the track prediction model to obtain a traffic track prediction result.

The invention discloses a manifold confrontation field self-adaption-based tumor diagnosis method, which comprises the following steps of: firstly, pre-training, and using a model pre-trained by source domain data as a model initialization parameter; then, manifold confrontation field self-adaption is carried out, a confrontation learning network framework is designed, a discriminant manifold propagation method is introduced to adjust structural parameters of a manifold module, manifold learning is used for assisting confrontation learning to further minimize the source domain and target domain feature mapping distance, and a target domain feature extractor is trained; and finally, testing, inputting a target domain test tumor image into a feature extractor after adversarial training to extract features, and then classifying data by using a trained classifier to obtain a tumor image classification result. The objective of the invention is to provide a manifold confrontation field-based adaptive tumor diagnosis method, which solves the dependence of existing deep learning on a large amount of labeled data, reduces the sample labeling cost in tumor diagnosis, reduces the labor intensity of doctors and experts, and realizes automatic and high-precision diagnosis classification of non-labeled medical tumor images.

The invention provides a lithium ion battery health state evaluation modeling method based on transfer learning. The lithium ion battery health state evaluation modeling method comprises the following steps: determining performance parameters of a lithium ion battery energy storage system having a change trend in a life cycle; taking the discharge capacity of the lithium ion battery energy storage system as a health state parameter, dividing the state parameter into four levels and setting corresponding labels; the method comprises the following steps: acquiring performance parameter data and label data of a lithium ion battery energy storage system in development and test stages, and constructing a source domain data set FS; using the source domain data set FS to train a mapping relation model between six performance parameters and state tags established by using a stacked auto-encoder algorithm to obtain a model parameter set; and constructing a target domain data set FT, optimizing the model parameter set, verifying the precision of the mapping relation model, and obtaining the lithium ion battery health state evaluation model. The technical problem of performance information fusion of multiple performance parameters in the use process of the lithium ion battery energy storage system can be solved, and health state grading of the lithium ion battery energy storage system is achieved.

The present invention provides a bionic robot fish control method and system based on fish body wave optimization, belonging to the field of robot control technology. The method comprises the following steps: obtaining the swing angle and swing angular velocity of the head of the bionic robot fish, as well as the joint angle and joint angular velocity of the tail in real time; generating the expected swing angle of the head of the fish, and the expected angle and expected angular velocity of each joint in the tail according to the preset expected trajectory of the bionic robot fish; obtaining the dynamic parameters of the bionic robot fish in the fluid; establishing an optimal control model, and obtaining the current optimal joint acceleration of the tail of the bionic robot fish as the control amount of the bionic robot fish. The method proposed by the present invention can effectively reduce the swing of the fish head during the swimming of the robot fish, and is more robust to parameter changes. When the parameters of the bionic robot fish change, stable control can still be achieved, and migration application can be achieved by simply modifying the joint parameters.

The invention relates to the technical field of non-standard automation equipment control, in particular to an industrial robot machining program migration method based on a configuration description file, and solves the problem that a non-standard industrial robot cannot migrate a machining program. The method comprises the steps that S1, a configuration description file of a prototype industrial robot is constructed, and based on object-oriented abstract modeling, kinematics, peripheral and communication modeling and a machining program are included; s2, the target robot is assembled according to the prototype physical structure, and it is ensured that the same configuration, component assembly, the axis movement direction and the origin coordinate are consistent; s3, loading the configuration file to the target robot; s4, correcting assembly errors through visual sensing, obtaining accurate coordinates among the parts, and then generating joint coordinates through inverse solution; and S5, the target robot executes the adaptive machining program, and program migration is achieved. According to the method, prototype machine information is packaged through the configuration description file, after the target machine is loaded, errors are corrected only through visual calibration, rapid adaptive machining can be achieved through kinematic chain inverse solution, and rapid copying of robots of the same configuration is achieved.

The present invention relates to the technical field of Java source code migration, and discloses a code migration method based on bytecode dynamic transcoding technology, including: constructing a code migration system, the system including a Java virtual machine and a Java migration unit, the Java migration unit including: a Java segmentation and encapsulation module, a Java migration module and a Java recovery module, any module in the Java migration unit has data interaction with other modules; and using a strong migration scheme, after the Java virtual machine determines the execution state of the Java program, the Java migration unit completely migrates the Java source code, data and process state information from the source node to the target node through a series of execution steps such as encapsulation, migration and recovery. The present invention enables the Java program to realize process migration between different nodes on the basis of Java source code integrity migration.

The embodiment of the present invention discloses a data migration method and device, which relates to the field of computer technology. A specific implementation of the method includes: constructing a data migration task, the data migration task indicates a target database; obtaining at least two data to be migrated from a source database; dividing the at least two data to be migrated into blocks to form at least two data blocks; according to the at least two data blocks, using one or more threads corresponding to multiple instances in the distributed cluster to execute the data migration task, so as to migrate the data to be migrated corresponding to the at least two data blocks to the target database. This implementation adopts a distributed cluster with multiple instances and multiple threads to execute the data migration task, thereby realizing high-concurrency and high-throughput data migration, thereby shortening the migration time and improving the migration efficiency.

The embodiment of the invention provides an automatic electronic fence setting method and device for protection device detection, and the method comprises the steps: capturing a real-time image of a target region through a camera, judging whether the real-time image of the target region is a black-and-white image or not through a protection facility detection model, and judging whether a moving target exists or not, inputting the real-time image of the target area into a YoV11 deep learning detection network under the condition that the real-time image of the target area is not the real-time image of the target area; detecting a protection facility in the real-time image of the target area through the YoV11 deep learning detection network, and outputting a current protection facility detection result; and taking the current protection facility detection result as a basis for setting an electronic fence algorithm, calculating an electronic fence coordinate parameter, and performing electronic fence setting according to the electronic fence coordinate parameter.

The application belongs to the technical field of fault diagnosis, and particularly relates to a twin data multi-modal fusion transfer diagnosis method for gear fault, which comprises the following steps: obtaining three mode components of Hilbert envelope spectrum, autocorrelation time domain waveform and autocorrelation envelope spectrum of gear measured and simulated signals, and respectively constructing three source domain subsets and three target domain subsets; performing JMMD mapping alignment on the source domain subsets and the target domain subsets of the same mode; training three independent DBSCAN classifiers by using the JMMD mapping features of the three source domain subsets respectively, and performing pseudo-label labeling on the JMMD mapping features of the target domain same mode components; updating the target domain sample labels by using a Sugeno fuzzy integral decision fusion method; repeating the JMMD mapping and the Sugeno fuzzy integral decision fusion until the maximum iteration number is reached, and obtaining a fault recognition result; and the application is supported by simulation data, and can realize accurate recognition of gear fault without the guidance of measured label data, and has a good application prospect in the field of gear fault diagnosis.

The invention provides tear-resistant aramid nano insulation paper as well as a preparation method and application thereof. The tear-resistant aramid nano insulation paper is prepared from multi-scale compounded meta-aramid chopped fibers with high elongation and good rigidity and para-aramid nanofibers with a unique nanoscale structure, rich functional groups and excellent mechanical properties through secondary forming and hot pressing. The excellent film-forming property of the para-aramid nanofibers and the advantage that rich functional groups on the surfaces of the para-aramid nanofibers and the meta-aramid chopped fibers can easily generate a strong network cross-linked structure are fully exerted, and the novel aramid paper-based insulating material with high tearability and high dielectric strength is developed. The defects of stress concentration damage, poor dielectric strength and the like caused by easy tearing of the existing aramid nano insulation paper are overcome, and the aramid nano insulation paper can be used as an ultrathin insulation layer to be applied to the next generation of highly lightweight and integrated special electrical equipment to realize miniaturization, integration and structural weight reduction of the equipment.

The invention discloses a method for degrading organic matters in desulfurized water through dual catalytic oxidation, and relates to the technical field of desulfurized water treatment. Comprising the following steps: naturally settling desulfurized water, and taking supernate to obtain impurity-removed liquid; adjusting the pH value of the impurity removal liquid to 3-4, adding a nano-iron supported catalyst and persulfate, uniformly mixing, adding hydrogen peroxide, and carrying out an oxidation reaction to obtain oxidized produced water; electrifying the oxidized produced water for electrolysis while continuously stirring and introducing air, and obtaining electrolyzed produced water after the reaction is finished; and adjusting the pH value of the electrolyzed water to 2.0-2.2, reacting to form a sodium jarosite precipitate, and filtering to obtain a treating fluid and sodium jarosite slag. On the basis of a Fenton / persulfate oxidation system, dual catalytic oxidation degradation of the organic matters is realized, the refractory organic matters are further mineralized into carbon dioxide and water through an electrolytic reaction, and a large amount of Fe < 3 + > in the desulfurized water is removed in a manner of forming sodium jarosite precipitate; and ferric hydroxide colloid is prevented from being formed in the subsequent size mixing procedure, so that the content of Li2O in the leachate is increased.

The invention relates to a data migration method and system, and belongs to the technical field of data processing.The method comprises the steps that a first Master node is deployed in a first machine room, and under the condition that the first Master node is initialized, cluster information and metadata information of a second Master node are synchronized to the first Master node based on connection between the second Master node in a second machine room and the first Master node; under the condition of the first limited routing rule, deploying a coordination node, a data node and a first consumption cluster in the first machine room; and migrating the data of the data node in the second machine room to the data node in the first machine room under the condition of carrying out rollover operation on the index. According to the method, the migration of the large-scale cluster is realized, the migration cost is reduced, the complexity of technical implementation is reduced, and the influence on the outside is minimized.

The embodiment of the invention discloses a style conversion method and device for a three-dimensional model. The method comprises the steps of obtaining a Gaussian model corresponding to a three-dimensional scene and color distribution of a target color style in a first color space; performing color decoding on the Gaussian model to obtain first color distribution of each Gaussian kernel in a first color space; transforming the first color distribution of each Gaussian kernel in the first color space by using the color distribution of the target color style in the first color space to obtain the second color distribution of each Gaussian kernel in the first color space; and encoding the second color distribution of each Gaussian kernel in the first color space to obtain a Gaussian model corresponding to the target color style. The color distribution of the Gaussian kernel is transformed by using the target color style, and an automatic color style conversion method for the three-dimensional scene is provided.

The invention discloses a heterogeneous data migration method and a related device, and relates to the technical field of data migration, and the method comprises the following steps: firstly, based on obtained data structure information of an original system and a target system, constructing a field corresponding relation between the original system and the target system; extracting the data structure information of each field in the original system, the data structure information of each field in the corresponding target system, the data volume of each field in the original system and the data storage space of each field in the corresponding target system; based on the extracted information, whether the data of the original system can be migrated to the target system or not is judged, factors of data structure difference and data size are considered, migration feasibility is ensured, format conversion is conducted on the data of the original system to meet the data structure requirement of the target system, the data obtained after format conversion is imported into the target system, and the data migration efficiency is improved. And heterogeneous data migration is realized.

This specification provides a training method, text processing method, and system for a large language model, comprising: obtaining a sample set, the sample set including multiple sample texts; obtaining a first large language model to be trained, the first large language model employing a linear attention mechanism; reusing at least some model parameters of the first large language model from pre-trained model parameters of a second large language model; the second large language model employing a non-linear attention mechanism; training the first large language model using the sample set; and obtaining a trained target large language model. This method can accelerate the training process, reduce computational and time costs, and improve the performance of the target large language model. Furthermore, it can optimize the computational complexity of attention to linear levels and optimize storage space.

The present disclosure relates to the technical field of speech synthesis, in particular to a speech synthesis method and device, electronic equipment and storage medium, the method comprising: obtaining text features of a target language and an identifier of an original language; performing style prediction of the original language based on the text features of the target language and the identifier of the original language to obtain style features, and querying a codebook corresponding to the target language based on the style features to obtain vectorized style features, the codebook corresponding to a language one by one and the codebook being used for vectorizing style features; and performing coding and decoding processing based on the vectorized style features to determine target speech of the target language. Different codebooks are used to vectorize style features for different languages, which can use the codebook of the target language when performing cross-language style transfer to alleviate the accent phenomenon.

The embodiment of the application discloses a data cross-database migration method and device, relates to the technical field of big data, and comprises the following steps: when a cross-database migration instruction for migrating target data in a first database to a second database is received, a currently idle process group is determined, wherein each process group comprises a first process and a second process, the first process is used for exporting data in the first database, and the second process is used for importing data into the second database; a first process in the currently idle process group is called to export the target data in the first database and store the target data into a preset transfer module; and a second process in the currently idle process group is called to import the target data stored in the transfer module into the second database. The application has the beneficial effect of simply and efficiently migrating data across databases.

The present application relates to the technical field of silicone key conductivity test, and discloses a silicone key conductivity reinforcement learning test system and method, wherein the silicone key conductivity reinforcement learning test method comprises the following steps: constructing a silicone key conductivity test environment; establishing a test knowledge base, extracting general feature representation of key conductivity performance through feature mapping and domain adaptation algorithm; constructing a test parameter optimization model, modeling the test process as a Markov decision process, and learning the optimal test strategy based on a reward function; adjusting the test parameters in real time during the test process, performing reliability analysis after obtaining the test results, and feeding back the new test experience to the knowledge base to realize the accumulation and reuse of test knowledge; the present application realizes the self-adaptive optimization of test parameters and the cross-key knowledge transfer by fusing reinforcement learning, transfer learning and meta-learning technology.

The invention discloses a catalytic cracking catalyst with high thermal wear resistance and a preparation method thereof, the catalytic cracking catalyst comprises the following raw materials: a zeolite molecular sieve, a matrix and a binder, the matrix is composed of a gamma-alumina precursor with a peptizing index less than or equal to 50% and clay, the binder comprises a phosphorus-containing compound, an aluminum-containing compound and a binder modifier; the binder modifier comprises a polymeshed chelating ligand, a metal center ligand, alcohol and organic acid, and the metal center ligand contains at least two metal ions; the catalytic cracking catalyst is prepared in a high shear emulsifier. Compared with the existing catalytic cracking catalyst, the catalytic cracking catalyst containing phosphorus and zeolite provided by the invention has the advantages of higher abrasive resistance, large pore volume, high molecular sieve content, high solid content and excellent reaction performance.

The present invention discloses a cross-region trajectory prediction method and system based on class-granularity spatiotemporal migration, and relates to the technical field of spatiotemporal data mining in intelligent transportation systems. The method comprises the following steps: collecting traffic trajectories of vehicles, and classifying the traffic trajectories according to sampling timestamps to obtain source area data and target area data, wherein the source area data includes source area historical data and source area future data, and the target area signal includes target area historical data and target area future data; constructing an initial trajectory prediction model, and training the initial trajectory prediction model using the source area data and the target area data to obtain a trajectory prediction model, wherein the initial trajectory prediction model includes: an embedding layer, a class-granularity spatiotemporal module, a distribution transformer and a de-embedding layer; collecting real-time traffic trajectories, and using the trajectory prediction model to perform traffic trajectory prediction to obtain traffic trajectory prediction results.

The present invention relates to the field of data processing technology, and discloses a snowfall calculation method, device, equipment and storage medium. By acquiring a geographic environment parameter set and snowfall associated parameter information of a target monitoring area, a geographic parameter feature map is constructed using the geographic environment parameter set, and a plurality of migration reference areas are generated in a geographic environment database. An independent reference snowfall prediction model is constructed according to the historical snowfall information of each migration reference area. Then, the reference snowfall prediction model is migrated based on model weight mapping, and the migration strategy is adjusted based on the model optimization period to achieve model migration and optimization between the migration reference area and the target monitoring area. On the one hand, the accuracy of snowfall calculation in the monitoring area under data scarce scenarios is ensured, and on the other hand, the degree of adaptation between the snowfall prediction model and the target monitoring area is improved, so as to avoid the problem of model incompatibility caused by the model migration process, and the model has good scene adaptability.