61 results about "Model synchronization" patented technology

Providing synchronization of data between application instances that execute script, and in preferred embodiments, without the use of synchronization primitives in the script and without the ability to reschedule events in at least one of the instances. Blocking code is provided in the application instances that is adapted for checking a flag to be set by the other component. One component can thereby inform the other component when it is ready to receive data, and the other component can inform the one component when the other component is finished with the data. In one scenario, the provided synchronization enables a browser instance that is terminating to safely provide its data to a separate browser instance, and enables the separate browser instance to know when the provided data can safely be manipulated.

One aspect provides an optical communication system. The system includes an optical-to-digital converter, a frequency estimator and a symbol synchronizer. The optical-to-digital converter is configured to receive an optical OFDM bit stream including an OFDM symbol bearing payload data and a symbol header preceding the OFDM payload data. The frequency estimator is configured to determine a carrier frequency offset of the payload data from the symbol header. The symbol synchronizer is configured to determine a starting location of the payload data within the bit stream by cross-correlating a synchronization pattern within the symbol header with a model synchronization pattern stored by the symbol synchronizer.

The embodiment of the invention provides a method for generating a heterogeneous database report, which is used for solving the problem that because the prior art can not support the connection of heterogeneous databases or the single connection type consumes resources, report data can not be acquired from service scenes of multi-dimensional application. The method comprises the following steps: establishing database connection with each heterogeneous database, and saving the database connection in the database connection pool of each heterogeneous database as a database connection object; acquiring data in each heterogeneous database through the database connection pool of each heterogeneous database; and integrating the data in each heterogeneous database into a data model which can be recognized by the report, and then, showing the data. The invention can efficiently acquire the data in each heterogeneous database, save the server resources and reduce the system overhead, and the integration of the data in each heterogeneous database can also enable the data to be synchronously shown in the report through the same model.

Disclosed is an invention, which cooperates with CAD for production design. It is characterized in that: employing the separate transmission mechanism of model synchronization and operation synchronization between client and server to realize real time synergetic design; also employing stream sockets for information exchange. For three-dimensional model, server terminal and each client terminals are provided with: model display module based on VRML files, customer interchange and communication modules; for two-dimensional model, the server terminal and client terminals are provided with communication module, remark module based on VECT library and communication module. The invention has the advantage of improving the real time communication efficiency between designers, of shorting production-developing cycle.

The invention discloses a dual-system-based graph-model data management method for a power distribution network. The method comprises the following steps of firstly, creating a debugging system and an operating system, before graph-model data are published, starting a model checking tool, and sequentially performing static verification and dynamic verification on a model in the current debugging system; secondly, after the model passes the verification, starting a model publishing tool, and synchronizing graph-model solid data of a commercial library layer according to a model export / comparison technology based on a fragmentation technology, a model synchronization technology based on a model cluster and a remote service access technology; thirdly, performing real-time library row-level dynamic updating in a messaging mode to achieve the seamless soft switching of side models of the operating system, and backspacing the graph-model data of the debugging system by virtue of the reverse synchronization of the graph-model solid data. According to the method, the correctness of the model can be guaranteed to the greatest extent, the real-time operating system cannot be influenced in the model publishing process, non-disturbance model updating is realized, and the operating reliability and the operating continuity of an on-line system are guaranteed.

The invention relates to an input method realized through network assistance and an input system. The language model base is maintained by using a network server and input is finished by calling the corresponding language model base by a user. The system comprises at least one piece of user equipment, the network server and a language model synchronizing device, wherein the user equipment is configured to be used by a user to enter characters and is configured to establish a local language model base according to the user input; the network server is configured to establish a network language model base on the basis of the received at least one local language model base and statistical analysis on other language materials; and the language model synchronizing device is configured to synchronize the local language model base and the network language model base. The invention also relates to the input method realized by using the system.

The invention provides a software model synchronization method based on code modification. By means of the method, non-substantial modification, non-essential modification and non-correlation modification codes are filtered out from changed codes, and modification codes relevant to a software model are obtained. The software model is matched with the relevant modification codes, a corresponding relation between elements in the model and elements in the codes is built, the obtained code modification information and the corresponding relation between the elements in the model and the elements in the codes are translated into updating operation, a new model is generated, and model synchronization is achieved. The software model synchronization method can be integrated in a code version management system, automatic or semi-automatic updating of the software model can be conducted by pre-defining rules, and consistency of the software model and the codes is guaranteed.

The invention discloses a method for synchronizing an AUTOSAR (automotive open system architecture) model to a Simulink model. The method includes: using an ARXML (AUTOSAR extensible markup language) source file to generate the AUTOSAR model, analyzing AUTOSAR object elements to obtain elements of the model, mapping the obtained elements of the AUTOSAR software module into the Simulink model to obtain a corresponding model port label set, and loading by means of HashMap; accessing to a generated folder of the Simulink model to search a journal file generated last time, and analyzing the journal file to obtain a corresponding Simulink model port label set of an MATLAB (matrix laboratory) / Simulink model; and comparing the obtained NEW label set with an OLD label set to obtain a result pool, and realizing model synchronization by means of Matlab orders. The method for synchronizing the AUTOSAR model to the Simulink model has the advantages of simplicity and convenience in conversion, low time consumption and low development cost.

The invention relates to a PaaS hybrid cloud construction method based on models. Specifically, the method comprises a step S1 of constructing a uniform model, describing resources, service types and management functions shared by PaaS platforms in cloud platforms, and managing the heterogeneous cloud platforms through a uniform view; a step S2 of constructing a cloud resource model, shielding heterogeneity of cloud platform management interfaces, describing function information of cloud platform managed modules and calling methods of the cloud platform management interfaces, and managing the cloud platforms in a model layer; and a step S3 of constructing model synchronization middleware, defining element mapping relationships from the uniform model to the cloud resource mode, automatically generating corresponding model conversion programs, and realizing a synchronization relationship between the models. According to the method, a management program can be established on the basis of the PaaS platform uniform model, and therefore, the difficulty and complexity of developing a PaaS hybrid cloud management system can be reduced to a great extent.

The invention discloses a method and a device for realizing sharing of a contextual model, so as to solve the problem in the prior art that user terminals can not share the contextual model. The method comprises: a sharing server receives a contextual model sharing request transmitted by a first user terminal, wherein the contextual model sharing request carries contextual model information requested for sharing and information of other users sharing the contextual model; and the sharing server transmits a contextual model synchronization request to user terminals corresponding to the information of the other users according to the contextual model sharing request, wherein the contextual model synchronization request carries contextual model information, so that the sharing server requests the user terminals to set the contextual model for communication according to the contextual model information. By adoption of the technical proposal, the method and the device can realize contextual model sharing between the user terminals and make the contextual model more flexibly set by the users.

The invention discloses a component business model for developing the application of a modeless documental database. The component business model comprises a modeless data storage layer business component, a business logic business component and a user interface layer business component. The modeless data storage layer business component comprises a collection component and a document attribute component. The business logic business component comprises a data operation business logic component, a page orientation business logic component and a code weaving business logic component. The user interface layer logic component comprises a window component, a unit component and an element component. The problem that the modeless database application software is low in developed efficiency is solved. The component business model is directly operated to replace the traditional fussy processes of software debugging, releasing and deployment. Developers only need to describe and model system businesses by the business component of the modeless database, Web page files are obtained by model conversion and model synchronization, and the Web page files, the business component and a model execution engine finally form Web application software executable.

The invention belongs to the technical field of software development and particularly relates to a universal software product line domain model and application model synchronization method. According to the method, the expression and the synchronization operation of domain models and application models of different types are abstracted in advance, a universal meta model is concluded through the domain models, and a set of fusion and transmission rules aiming at model changing based on the meta model is concluded through the application models. On the basis of the universal meta model and the universal synchronization rules, a universal model synchronization mechanism is provided for the domain models and the application models of different types in a software product line, the mechanism supports automatic model changing identification, fusion and transmission, and meanwhile corresponding synchronization operation options are provided for a domain engineering team and an application engineering team of the software product line in a synchronization process. When the method is used, only a model converting program between a model of a certain type and the universal expression form of the model needs to be compiled, and evolution synchronization aiming at the newly-introduced model of the certain type can be supported.

The invention relates to a model-based mixed cloud construction method. Element mapping relations between a unified model and a cloud resource model are set via a model synchronous middleware, so management operation on the unified model is automatically switched to the corresponding management operation on the cloud resource model and affects cloud resources in a cloud platform, so relation synchronization between models is finished. The unified model carries out description via shared resource types and management functions of the cloud platform. The cloud resource model is used for shielding heterogeneity of management ports of the cloud platform. The model synchronous middleware is used for mapping the unified model on the cloud resource model. According to the invention, management programs can be established on the basis of the unified model of a cloud computing platform, so degree of difficulty and complexity for developing a mixed cloud management system are reduced.

The invention discloses a block chain cross-chain-based federated learning method and device, and the method comprises the steps: carrying out the training of a federated learning model through local data in a single block chain network, and transmitting model parameters to a model aggregation smart contract through a private transaction, thereby achieving the aggregation of different node model parameters in a single chain, synchronizing model parameters at each node; the model aggregation smart contract sending the latest model parameters to a cross-chain network through cross-chain private transaction to realize model synchronization between different chains. In the whole process, data of a single network node and data of a cross-chain network node are not exchanged, it is ensured that model parameters of each node are not leaked through a private transaction mode, training of different node data of different block chain networks on a federated learning model is achieved while data privacy security is ensured, a trained data set is expanded, and the accuracy of the model is improved. An integral mechanism is adopted to improve the enthusiasm of each member to contribute to a data training model, and the model training effect is further improved.

The invention discloses a method and system for correcting digital twinborn model data of an evaporation area of a thermal power plant, and the method comprises the following steps: obtaining the operation physical data of a physical system of the evaporation area, carrying out the estimation of the physical quantity of the evaporation area through a self-adaptive square root unscented Kalman filter, and obtaining the real operation physical data; using the mechanism model and a neural network to construct and obtain an evaporation zone digital twinborn model; and synchronously operating the evaporation zone physical system and the evaporation zone digital twinborn body model to obtain real operation data and twinborn data, and correcting the twinborn data based on the real operation data. According to the method and system, the adaptive square root unscented Kalman filter, the neural network and other methods are combined, and real-time correction of the model and data of the digital twinborn in the evaporation area can be achieved, so that the consistency simulation of the digital twinborn on the running condition of the evaporation area is achieved.

Detecting timing of a synchronization sequence included in a data stream transmitted in a noisy channel. The synchronization sequence is a known data sequence purposely injected into the data stream for synchronization. A data stream waveform, is obtained from a noisy communication channel. The data stream waveform includes an instance of a synchronization sequence waveform. One or more samples of the synchronization sequence waveform are obtained. One or more samples of a model synchronization sequence waveform are obtained. The model synchronization sequence waveform models an expected waveform for the synchronization sequence being transmitted on the communication channel by applying the synchronization sequence to a model of the communication channel. The instance of the synchronization sequence waveform is correlated with the model synchronization sequence waveform by correlating one or more samples of the synchronization sequence waveform with the one or more samples of the model synchronization sequence waveform.

The invention discloses a heterogeneous system and a model synchronization method of the heterogeneous system. The heterogeneous system comprises a debugging system and an operation system, wherein identical assemblies are deployed on the debugging system and the operation system; and each assembly comprises a cooperated general control program and a plurality of sub programs called by the cooperated general control program. The model synchronization method of the heterogeneous system provided by the invention has the advantages that the synchronization of a whole model is completed through the communication management and the flow process cooperation on sub tasks in the heterogeneous system; the problems of communication management and flow process cooperation of different sub tasks of the heterogeneous system are solved, so that the model comparison and synchronization of the heterogeneous system can be realized; the system level data analysis can be conveniently realized; and the data sharing among the systems is expanded.

The invention provides a main-standby scheduling model data synchronization method for an electricity SCADA (supervisory control and data acquisition) system. The method includes that (1), when model data are changed, a main scheduling system sends a model synchronization request to a standby scheduling system; (2), when detecting the model synchronization request, the standby scheduling system receives the model synchronization request of the main scheduling system; (3), the main scheduling system writes the model data into a server of the standby scheduling system and performs model synchronization process tracking; (4), the standby scheduling system feeds back synchronization information to the main scheduling system, and the main scheduling system writes the synchronization information into a log after tracking the synchronization information. The main-standby scheduling model data synchronization method has the advantages that automatic synchronization of the model data of the main scheduling system and the standby scheduling system is achieved, and the model data, generated and stored on the basis of definition and instantiation, of the main scheduling system can be safely synchronized to a model database of the standby scheduling system in real time, so that simplicity and convenience in user maintenance are improved, maintenance cost is reduced, and date security is enhanced.

A computer-implemented method for collaborative editing of an electronic model of a design object may include (1) identifying clients configured to submit design operations to be performed on a shared design model, (2) receiving, from a first client, a first design operation to be performed on the shared design model prior to receiving, from a second client, a second design operation to be performed on the shared design model, (3) performing the first design operation prior to attempting to perform the second design operation, (4) identifying a conflict between the two design operations, and (5) in response to identifying the conflict between the design operations, notifying the second client that the second design operation was not performed on the shared design model. Various other methods and systems are also disclosed herein.

The invention relates to a model management method and system, and the model management method comprises a model publishing step which is used for obtaining a trained model, judging the publishing type of the model, and carrying out the model publishing through a model management platform according to the publishing type; a model registration step for registering published information of the modelto a distributed coordination service, the published information including but not limited to a model name, a model version, a file name, a storage type and a published type; and a model synchronization step for synchronizing instance information of a model in the distributed coordination service, the instance information including but not limited to the model name, the model version, the model state and the model request quantity. Through the application, model deployment is realized and simplified, the state information of each model can be visually acquired through the model management platform, the working efficiency is greatly improved, and the cost is reduced.

The invention relates to a medical artificial intelligence model training system based on a block chain. The medical artificial intelligence model training system comprises a model distribution module, a parameter collection module, a model training module, a model synchronization module, a block chain infrastructure, a medical artificial intelligence model library, a user information library, a model updating process library and a model publishing module. According to the system, the data islands of the medical institutions can be broken through, and the risk of private data leakage of the medical institutions is reduced. Only the model and parameters are transmitted in the system and thus the artificial intelligence model can carry out learning under more data samples under the conditionof private data protection, and the accuracy and generalization ability of the model are improved. In addition, the model training updating process in the system is traceable, so that the safety, credibility and controllability of the model are realized.

The invention provides an automatic code generation method, a code generator and a readable storage medium, which are used for solving the problems that secondary development is not supported, the code coupling degree is high and the flexibility is poor. The method comprises the following steps: firstly, configuring a data source, and storing the data source into a corresponding database to form adatabase table; creating a corresponding model package according to the configured data source; utilizing the model package to construct a service model; judging a model synchronization mode; synchronizing to a database, and reconfiguring a data source; performing model synchronization, and judging whether the service model is a relationship model or not; if the model is a singleton model, processing the model to generate a singleton model code; If the model is the relationship model, processing the model to generate a relationship model code.

The invention relates to a method and apparatus for distributed training of a machine learning model. The method comprises the following steps of: in a preset number of iteration steps of distributedtraining, obtaining the sequence of the current iteration steps, the preset number and the N being in a logarithmic relationship; based on the sequence, obtaining one node from the N nodes as a targetnode of the current node in the current iteration steps; communicating with the target node to obtain model information shared by the target node; updating the machine learning model of the current node according to the model information of the current node and the model information of the target node, so as to synchronously obtain the model information of the N nodes by the machine learning model of each node after completing the preset number of iteration steps. According to the method and apparatus of the embodiments, the convergence efficiency of distributed training can be ensured; meanwhile, the communication traffic of a single node can be reduced to S, and the effect of shortening the communication time is achieved; in addition, each node only shares information with its respective target node, so that the load balance among the nodes can be ensured.

The present disclosure relates to a method and apparatus for distributed training of machine learning models. The method includes: in the preset number of iteration steps of distributed training, obtaining the order of the current iteration step; the preset number has a logarithmic relationship with the N; based on the order, obtaining a node from N nodes as The target node of the current node in the current iteration step; communicate with the target node to obtain the model information shared by the target node; update the machine learning model of the current node according to the model information of the current node and the model information of the target node to complete the preset amount After the iteration step, the machine learning models of each node obtain the model information of the N nodes synchronously. In this embodiment, the convergence efficiency of distributed training can be guaranteed; at the same time, the communication volume of a single node can be reduced to S, achieving the effect of shortening the communication time; in addition, each node only shares information with its own target node, which can ensure that each node Load balancing among them.

The embodiment of the present invention discloses an AGC information synchronization method and system between heterogeneous scheduling automatic main and backup systems. According to the present invention, the AGC model synchronization of a main scheduling side and a backup scheduling side is realized by a model file, the operation information synchronization is realized by a universal message form, and the technical problem in the prior art that the main and backup systems are a heterogeneous system of different manufacturers by using a database and a manually operated message cross-system to synchronize, is solved. The method of the embodiment of the present invention comprises the steps of receiving the model file corresponding to a main scheduling side AGC model after the maintenance and the first verification; importing the main scheduling side AGC model corresponding to the received model file to a backup scheduling side AGC maintenance base, then carrying out the second verification on the main scheduling side AGC model in the AGC maintenance base and publishing the AGC model after the second verification to a backup scheduling side AGC operation base; receiving an operation message corresponding to the main scheduling side operation information and operating correspondingly according to the operation message.

An online model interaction method based on a power grid operation service bus and a public information model synchronizer is characterized in that a client end applies a newest network model to a power grid dispatching EMS system by the power grid operation service bus (OSB) in an online mode, and transmits the CIM model returned by the EMS system into a fast CIM model adapter, and the newest model data is finally guided into other dispatching subsystems by links such as CIM model analysis, model data classification, model data association, model increment screening and model attribute comparing.

The invention relates to a model-based industrial edge cloud cooperation system and method. The system comprises: an edge side, which are used for realizing industrial protocol conversion based on a unified protocol library, constructing an industrial model operation environment based on container operation, realizing industrial model attribute definition and method mapping based on model management, and analyzing and executing the industrial app and the industrial model constructed on the cloud side based on a model analysis engine; and the cloud side, which is used for performing industrial model synchronization with the edge side, realizing industrial model management and constructing an industrial APP. According to the method, the business logic is compiled on the cloud side based on the model, the industrial app is generated, the industrial app is issued to the edge side through model management, the edge side analyzes the industrial app based on the model analysis engine and the data association engine, execution is performed through the model management module, and the industrial app can be rapidly constructed through the cooperation method based on the analysis.