150 results about "Dual network" patented technology

The invention provides an anomaly detection and maintenance decision optimization method and system for automatic test equipment, and relates to the technical field of automation, and the method comprises the steps: deploying an edge calculation unit and a multi-level sensor network to collect equipment data, carrying out the fault diagnosis through a deep network model, and constructing a maintenance decision model in combination with equipment operation data. The model is input into a quantum enhancement dual-network system to calculate a reward value, and is used for training a maintenance strategy network under a federal reinforcement learning framework. And generating a maintenance resource configuration scheme through a quantum evolutionary algorithm, and generating a maintenance scheme through optimization of the multi-criterion evaluation model and the self-adaptive multi-target differential evolutionary algorithm. According to the scheme, a high-fidelity digital twin model and a multi-task deep transfer learning network are input for verification and updating, updated data are transmitted to a field control system through an industrial Internet of Things platform, and the updated data are recorded by using a block chain technology for updating a deep network model.

The invention provides an automatic filling and reviewing and knowledge base integrated optimization method based on a dynamic rule engine, and relates to the technical field of software engineering, and the method comprises the steps: carrying out the standardization processing of electronic form data, and generating a to-be-reviewed data set based on a preset filling rule. And the constructed dynamic rule engine realizes time sequence tracking and self-adaptive recombination of the rule, a scenarized rule chain is generated for rule matching, and finally an examination result is obtained. According to the method, through multi-level knowledge modeling and knowledge dual network construction, potential knowledge conflicts are detected, the repair strategy is generated, and the accuracy and efficiency of filling and examination are remarkably improved.

The invention relates to the technical field of polyester production control, in particular to an intelligent control method and system for a PET short fiber traction-cutting production line based on a PLC, a storage medium and a program. According to the method, a dual-network architecture in which a high-speed bus and an information bus cooperate is constructed, ARX tension prediction, a Kalman observer, resonance notch, energy efficiency constraint model predictive control (e-MPC) and two-degree-of-freedom speed feed-forward compensation are fused, and high-precision tension control and resonance suppression of the traction-cut-off linkage system are achieved. The method supports redundancy switching and self-calibration of the tension sensor, has millisecond-level response and + / -1.5% steady-state fluctuation control capability, remarkably improves control stability and energy-saving level, and is suitable for intelligent upgrading of a high-performance PET short fiber continuous production line.

According to the incineration pollution cooperative control method based on multi-target reinforcement learning, a multi-modal sensor array is deployed in a hearth, a flue and a fly ash processing unit, multiple parameters such as combustion temperature and flue gas retention time are collected in real time to construct a high-dimensional dynamic characteristic matrix, and data collection and processing are carried out to meet subsequent model requirements. Based on the dynamic characteristic matrix, defining three types of objective functions of pollutant emission minimization, fly ash metal recovery rate maximization and resource cost minimization, adopting a depth deterministic strategy gradient algorithm to construct a multi-objective reinforcement learning joint optimization model, and taking a combustion air ratio and the like as an action space; and strategy exploration and value evaluation are realized through a dual-network structure. Meanwhile, a priority evaluation module based on fuzzy logic is constructed, the weight coefficient of an objective function is dynamically adjusted according to real-time working conditions, and dynamic weight distribution and multi-objective collaborative optimization control are achieved. According to the method, multiple targets in the waste incineration process can be effectively balanced, the incineration efficiency and the resource utilization level are improved, and pollutant emission is reduced.

The invention discloses a voltage regulation and control method and system suitable for a distributed energy storage photovoltaic transformer area of a power distribution network. The method comprises the following steps: establishing a transformer area active power distribution network model for carrying out voltage regulation and control by considering distributed photovoltaic, wind power, an energy storage system, a static var compensator and a load node; a voltage regulation problem is modeled as a Markov decision process, a state space comprises multi-source power information and energy storage state information, and an action space is controlled quantities of a static var compensator and an energy storage system; a Markov decision is solved by adopting a depth deterministic strategy gradient algorithm based on Actor-Critic dual networks, a control action is generated through a strategy network Actor, and a current strategy is evaluated through a value network Critic for training; and inputting the trained strategy network based on the current observation state, outputting an optimal energy storage active / reactive power output and SVC reactive power compensation control instruction, realizing online adjustment of node voltage, and realizing stable control of transformer area voltage.

A satellite adaptive modulation method based on deep reinforcement learning: 1) Perform initialization operations; initialize the state space, action space, and greedy parameter; 2) The signal receiving end on the ground receives the signal from the satellite downlink and extracts the pilot information in the current frame for signal-to-noise ratio (SNR) estimation. After the receiving end calculates the SNR estimation result, it is transmitted to the ground sending end through the ground feedback link; 3) The sending end translates the selected action into the corresponding modulation method and coding rate; 4) Determine whether the current iteration number is an integer multiple of the preset network update step number. If so, go to step 5) for network update; if not, update the SNR state and return to 2) to enter the next round of iteration; 5) Introduce the concept of a dual network and improve the learning effect and accelerate the convergence of the results by optimizing the structure of the neural network. 6) Update the SNR state, increment the greedy parameter, and return to 2) for the next round of iteration.

The invention discloses an electric vehicle path planning method based on evolutionary reinforcement learning in a dual-network fusion scene, which is based on each component of an electric vehicle path planning problem, and comprises the following steps of: carrying out mathematical modeling on the electric vehicle path planning problem by a vehicle dispatching center, a transportation task point and a charging station, and establishing various constraints conforming to reality; according to the method, a target function with the purpose of reducing energy consumption and time consumption of a vehicle for completing a transportation task is obtained, and then an improved evolution reinforcement learning optimization algorithm is adopted, the determined target function is used as an optimization target, and on the premise that various constraints meet and the transportation task is completed, a path with the minimum energy consumption and the shortest time consumption of the electric vehicle is calculated. According to the method, convergence of a loss function is accelerated by utilizing a double-layer replay buffer area structure, and a mode of selecting empirical data in different periods is standardized and trained by designing adaptive parameters, so that the defects of long optimization time consumption, low search efficiency, easiness in falling into local optimum and the like in solving an electric vehicle path planning problem in the prior art are overcome.

The invention discloses a data consistency verification method for dual-network communication of a digital substation, and relates to the technical field of operation data monitoring. The digital network A and the digital network B are arranged, time correction pulses are sent to the digital network A and the digital network B, then time synchronization of the digital double networks is completed, a plurality of collection cores are arranged in the digital network A and the digital network B, multiple fault-tolerant mechanisms are set for the collection cores, and the time synchronization of the digital double networks is completed. The message data in the operation process of the transformer substation is acquired through the acquisition core, the acquisition process is monitored through a multiple fault-tolerant mechanism, structured analysis is performed on the message data, and then a corresponding SV message detection data table and a GOOSE message detection data table are generated; and sequentially comparing the SV message detection data tables of the digital network A and the digital network B with each channel data in the GOOSE message detection data, and judging an abnormal digital network according to a comparison result.

The invention discloses a home appliance dual-network hub site selection planning method based on a hybrid heuristic algorithm. The method comprises the following steps: firstly, modeling a home appliance dual-network hub site selection planning problem into a mathematical model taking the minimum comprehensive cost as an optimization target; and then, solving by using a hybrid heuristic algorithm. The algorithm is a priority competition allocation algorithm based on a simulated annealing framework. Two attribute indexes of a comprehensive load coefficient and a distance sensitivity coefficient of a site are defined, the competition degree of the site for hub resources and hub distance sensitivity of the site are quantified, a reasonable distribution sequence is determined, resource conflicts are reduced, efficient configuration of hub resources of a sales service network is achieved, and the service efficiency of the sales service network is improved. The problem that a traditional clustering method is low in solving quality is effectively solved.

The invention relates to the technical field of energy storage control systems, and discloses a networking structure of an energy storage control system and a control method, in the networking structure, a plurality of main control modules are in communication connection in sequence, the first main control module and the last main control module are in communication connection with an EMS system, and hand-in-hand ring network communication is achieved; the plurality of main control modules and the EMS system are respectively connected to the switch through the Ethernet, and Ethernet communication is realized through the switch. According to the control method, a host can carry out sorting and distribution according to the electric quantity and the number of all current battery compartments, charging is started from the battery compartment with the minimum SOC during charging, and discharging is started from the battery compartment with the maximum SOC during discharging. According to the invention, a dual-network dual-redundancy communication mode between the main control module and the EMS system is adopted, the number of control equipment is reduced, the software and hardware cost is reduced, the continuous stability of system operation is ensured through a redundancy communication structure, and the number of operation equipment can be adjusted and the operation loss can be reduced according to full and full discharge conditions.

The invention discloses a carrier rocket avionics system based on a TSN (Time Sensitive Network) and a carrier rocket avionics data transmission method, and belongs to the technical field of carrier rocket control. A network is physically or logically divided into a periodic time-sensitive data stream, a high-speed time-sensitive data stream and a best effort data stream, the three data streams are subjected to isolation management, and transparent interconnection between a 1553B bus network and a TSN switching network is realized on a protocol layer through an intelligent protocol conversion gateway. Meanwhile, in a network management layer, through an instruction response type configuration protocol, common Ethernet equipment is seamlessly incorporated into the global scheduling and control category of the TSN network. According to the invention, by constructing the avionics network system with three-layer partition and dual-network fusion, multi-network seamless fusion on the rocket can be realized, the real-time performance of the network is improved, and the key requirements of rocket recovery are met.

The invention discloses an RDMA (Remote Direct Memory Access) dual-network-card active-standby switching method and equipment, and the method comprises the steps: intercepting an RDMA call of an application program through a built-in RDMA function library of a middle library, and if the RDMA call is intercepted, responding to the RDMA call through the built-in RDMA function library to carry out RDMA communication, the method comprises the following steps: mapping resources required by RDMA communication into resources of a middle library so as to enable an application program to directly access the resources of the middle library, monitoring the change of an RDMA network card through the middle library, and performing secondary reconnection by using a standby RDMA network card when monitoring that a main RDMA network card has a fault; and after the reconnection is successful, switching the resource of the middle library of the associated main RDMA network card to the associated standby RDMA network card so as to continue the RDMA communication. The invention aims to solve the problem that communication cannot be taken over when the network cards fail in the RDMA technology, and quick and stable main-standby switching is realized for the RDMA dual network cards.

The invention discloses a novel power distribution communication network evaluation method and system under extreme disasters, and the method comprises the steps: constructing a dual-network coupling model of a novel power distribution system 10kV line, a power grid node failure model under the extreme disasters, and a communication node failure model after the power grid node energy supply is lost under the extreme disasters; correlation characteristics of a power distribution system power grid and a communication network are fully considered, failure conditions of power grid nodes and communication nodes are accurately judged, probability distribution of wind speed flowing through the power grid nodes under extreme disasters is generated, wind speed conditions are truly simulated, and a basis is provided for judging failure of the power grid nodes. Cascading failure simulation is carried out based on a dual-network coupling model of a novel power distribution system 10kV line, a power grid node failure model under an extreme disaster, a communication node failure model after power grid node energy supply is lost under the extreme disaster and probability distribution of power grid node flowing-through wind speed under the extreme disaster, and then the reliability of the communication network is evaluated. And the accuracy and practicability of reliability evaluation of the communication network are improved.

The application relates to a small-sample residual life prediction method based on task incremental meta-learning, which comprises the following steps: acquiring a multi-element sensor failure signal and an actual residual life of an airplane engine, and pre-processing and selecting the signal; constructing a dual network, including a meta network and an auxiliary network, wherein the meta network regards the mapping of sensor data of each device to the residual life as a regression task, and automatically learns a task-specific model of each device by using meta learning; the auxiliary network generates pseudo labels for unlabeled samples of non-failed devices; the dual network is trained based on labeled data of failed devices and unlabeled data of non-failed devices by using a task incremental training algorithm; the pseudo labels of test device samples are predicted by using the auxiliary network, the meta network is updated to a task-specific model based on a pseudo label support set, and the residual life of the test device is predicted by using the task-specific model. Compared with the prior art, the application can realize accurate residual life prediction under a small number of failure samples.

The invention aims to provide an encrypted traffic classification method and classification model based on a dual network byte traffic diagram, and belongs to the technical field of information security. According to the method, a position one-hot coding technology is innovatively applied to construction of vertexes of a byte traffic diagram, a position window is introduced to construct edges, position information is successfully integrated into a traffic diagram structure, and potential original features in traffic are mined; the method comprises the following steps: constructing a traffic diagram based on position codes of a head byte and a load byte of a data packet, extracting features of the head byte traffic diagram and the load byte traffic diagram, performing cross feature fusion on the two features, and finally completing classification of encrypted traffic by adopting a bidirectional LSTM (Long Short Term Memory) based on the time sequence of the traffic. A double graph and feature fusion mode is adopted, and the flow is represented more comprehensively in terms of semantics and space structures. Compared with a current encrypted traffic classification method, the method has higher recognition accuracy and can be applied to an actual generation environment.

The invention discloses a network implementation method, device and equipment for dual-network service coexistence based on an Android system, and the method comprises the steps: judging whether an accessed second network exists or not when a first network is accessed, and if the accessed second network exists, executing the operation of the first network; if it is judged that the accessed second network exists, the network score of the first network is determined according to the network score of the second network and a preset priority network, whether the direct connection routing information of the first network needs to be backed up or not is determined, and if it is judged that the accessed second network does not exist, whether the direct connection routing information of the first network needs to be backed up or not is determined; the current priority network is determined according to the network score, and the backup direct connection routing information is updated, so that the preset priority network set or preset in the controller by the user becomes the current priority network in a mode of intervening the network score of the first network, and therefore, the user can select the network as required, and the user experience is improved. The current priority network and the non-priority network can be interchanged through a network scoring mechanism.

The invention provides a cooperative control system for dual-network heterogeneous equipment, and the system comprises a master station which communicates with a relay station through a first industrial network; at least one substation communicating with the relay station over a second industrial network; wherein the master station and the sub-station carry out data synchronization through a data synchronization method. According to the invention, through the collaborative architecture of the master station, the substations and the relay stations, in combination with the dynamic protocol conversion and dual-network data mirroring technology, the problem of data synchronization among heterogeneous network equipment is effectively solved, meanwhile, the continuous operation of the system is ensured by using a hardware accelerated fault isolation mechanism, and the method has the effects of improving the reliability and flexibility of the system and reducing the downtime.

The invention discloses a port authentication and three-right function test method and device, equipment and a storage medium, and relates to the technical field of communication, and the method comprises the steps: generating a test message based on a preset authentication protocol and preset authentication information; sending the test message to a switch to enable the switch to complete port authentication and three-right function test; when the port state of the switch is changed, capturing an interaction message between the switch and a remote user dial-in authentication server; and obtaining a test report based on a machine learning model, the test message and the interaction message. According to the application, the closed-loop flow of authentication request sending, response capturing and result analysis is realized through division and cooperation of the double network cards, the first network card simulates the client and the second network card monitors the state of the switch and controls the monitoring server, and in the stage of result analysis, indexes such as authentication success rate, response delay and concurrent session capacity are counted through a machine learning algorithm, so that the result analysis efficiency is improved. And a multi-dimensional performance evaluation report is provided.

The invention provides a semi-supervised deep learning image restoration enhancement method based on dual-network cooperation. The method comprises the following steps: constructing a dual-task branch image restoration network, training the dual-task branch image restoration network by adopting a teacher-student semi-supervised training mode, and performing feature extraction and restoration enhancement processing on a to-be-restored image by a restoration branch in the trained dual-task branch image restoration network, outputting the restored image, the low-order semantic features and the high-order semantic features, fusing the low-order semantic features and the high-order semantic features by an evaluation branch, and obtaining a structural similarity index SSIM score matrix by using a self-attention mechanism; and according to the SSIM score matrix, determining the confidence coefficient of pseudo label screening in the training process of the teacher-student semi-supervised training mode of the dual-task branch image restoration network. According to the method, values of a large number of low-quality samples and a small number of high-low-quality paired samples are fully mined, high-quality pseudo labels are generated, and accurate restoration and enhancement of trackside images are realized.

The present invention discloses a method for handwritten Chinese character recognition and correction, including obtaining a handwritten Chinese character image and performing noise reduction processing on it; performing image segmentation on the handwritten Chinese character image after noise reduction processing to extract the image to be recognized with Chinese characters; establishing a training data set and a first recognition model for target handwritten Chinese characters; detecting the image to be recognized based on the trained first recognition model for target handwritten Chinese characters to obtain a first recognition result of the handwritten Chinese characters in the image to be recognized; correcting the scribbled fonts in the first recognition result based on a hypergraph neural network; the present invention uses an EfficientDet dual network to recognize images, fuses a feature pyramid model, processes the features of scribbled and non-scribbled handwritten Chinese characters, uses an adversarial network to generate a large number of adversarial scribbled sample data to improve the recognition effect of scribbled Chinese characters, and at the same time uses a hypergraph neural network to accurately identify the scribbled types of the recognized scribbled Chinese characters. Therefore, the speed of recognizing scribbled Chinese characters is fast and the accuracy of correcting scribbled Chinese characters is high.

The present invention provides a communication method and apparatus for a dual network interface card terminal, and an electronic device. The method comprises: calculating the network link quality of a first network and the network link quality of a second network, wherein the first network is a wireless network accessed by a first network interface card, the second network is a wireless network accessed by a second network interface card, and the network link quality comprises a signal strength indication; comparing the network link quality of the first network with that of the second network, and taking the network with superior network link quality as a target network; and switching the target network to a primary network, and taking the other network as a backup network, wherein the primary network is a network for data transmission. The present invention can select and switch between the primary network and the backup network on the basis of the wireless network link quality, and automatically select an ideal frequency band for communication, ensuring continuous and stable transmission of important data.

The application discloses a mobile cart for use in a hospital environment, and more particularly discloses a technique for equipping a mobile device with two different types of wireless communication devices, one of which is part of a mesh network, the mesh network may serve as a backup path to route data to the Internet. When the Wi-Fi connection fails or is weak, a mobile device, such as a mobile cart, may dynamically switch to a mesh network. This dual network approach enhances the reliability of data transmission, reduces the dependency on a single type of network, and ensures continuous operation, which may be important in certain environments, such as in a health care environment.

The present invention relates to a full-coverage optimal path planning method, comprising the following steps: Step 1: Obtain map information; Step 2: Establish a labeled double Petri net model; Step 3: Establish a mathematical model; Step 4: Establish an integer linear programming model; Step 5: Solve the integer linear programming model to implement a dynamic full-coverage path planning method. The robot optimal path dynamic planning method of the present invention can establish a labeled double Petri net model according to map information, record the horizontal or vertical movement corresponding to the transition through the label information of the transition, so as to observe whether the robot turns during movement.

A swelling-resistant hydrogel with excellent mechanical properties, and a preparation method and application thereof. The present invention belongs to the field of double-network hydrogels. The purpose of the present invention is to solve the technical problem that the existing hydrogels cannot take into account excellent swelling resistance, high strength and toughness, and electrical conductivity. The hydrogel of the present invention has an interpenetrating double network, wherein the chitosan molecular chain is used as the first network, butyl acrylate, 2-hydroxyethyl methacrylate and acrylic acid are used as monomers, and dimethylaminoethyl methacrylate maleate is used as a cross-linking agent to copolymerize to form a second network, and the first network and the second network form a double-network physical cross-linking structure with hydrogen bonding. The hydrogel prepared by the present invention has good mechanical properties, high continuous conductivity and swelling resistance. Based on these excellent properties, the hydrogel is expected to be used as a flexible wearable electronic device to broaden its application range.

Generally described, one or more aspects of the present application relate to providing a casting server that can pair specific user devices to appropriate corresponding casting devices and control the network communication between the user devices and the casting devices, such that only the casting devices or display devices that a given user is authorized to access are made visible to the user via the user's device. By doing so, the user is prevented from accessing or casting the user's content to other casting devices and display devices to which the user may not have access such as those in other users' rooms.

This application discloses a dual-network synchronous switch and a dual-network synchronous switching method, relating to the field of network communication technology. The dual-network synchronous switch comprises a field programmable gate array (FPGA) and a central processing unit (CPU). The FPGA includes a high-speed interface processing module, a time frame module, a MAC processing module, a clock synchronization module, a time maintenance module, an FPGA time interaction module, and a data exchange module. The CPU includes a data processing module, a CPU time interaction module, and an Ethernet MAC module. This application implements clock information exchange through the FPGA time interaction module and the CPU time interaction module, and data exchange through the data exchange module and the data processing module. The Ethernet MAC module encapsulates the received clock information and data, which can then be communicated with other networks after encapsulation, thereby achieving clock synchronization and data exchange between the dual networks.

The invention relates to the technical field of system software communication, and discloses a read-write method based on a dual-network redundant database. The method comprises the following steps: S100, configuring a first database connection pool and a second database connection pool when loading for the first time, and arranging the first database connection pool and the second database connection pool in an accessible first address and a second address respectively; s102, acquiring a database client from a database connection pool corresponding to the first address; s104, judging whether the acquisition is successful or not, if so, turning to S106, otherwise, turning to S110; s106, returning to the database client; s108, performing reading and / or writing operation on the database server side by using the returned database client side; s110, exchanging positions of the first database connection pool and the second database connection pool, and counting exchange times; and S112, judging whether the current count value exceeds the maximum number of times or not, if so, prompting that the database client is failed to obtain, and otherwise, returning to S102. Therefore, high-reliability reading and writing of the database can be realized.

The present invention provides a weakly supervised shadow detection method for mining complementary features through dual networks. First, the class activation map method is used to obtain the initial seed regions of shadows. Then, the region segmentation network further learns the region-based shadow characteristics based on the initial seed regions to estimate the shadow regions more accurately, and uses the shadow detection results of the pixel segmentation network to train the region segmentation network. Finally, in the process of mutual learning, the region segmentation network and the pixel segmentation network complete the fusion of region features and context features, realizing the accurate segmentation of shadow regions. The present invention can well complete the shadow detection task under the condition of using weak labels, greatly reducing the dependence on the high-level labels of the image dataset, effectively utilizing the features extracted by the classification network in superpixel blocks and the context features directly obtained by the segmentation network from the whole image, so that the purpose of shadow detection can be effectively achieved under weak supervision.

The application discloses a double-network redundancy transmission method and device based on link selection, and the method comprises the following steps: a data sending node duplicates message data and sends the message data to two independent parallel transmission networks through two network cards; when the data receiving node receives the message data sent by the data sending node from the corresponding transmission network through any one of the two network cards, the data receiving node checks a built-in node link state table, judges whether the record of the data sending node exists, if the record exists, judges whether the message data is sent according to the record of the data sending node, and updates the record of the data sending node in the node link state table according to the message data; if the record does not exist, the record of the data sending node is added in the node link state table, and the message data is sent according to the record of the data sending node; the application can construct two physically completely independent networks, and data can be transparently, efficiently and reliably transmitted in the two networks.

The present invention relates to the field of distribution network technology, and specifically to a dynamic two-layer optimization scheduling method for an electrified transportation-distribution network, comprising: constructing an objective function of a distribution network optimization scheduling model with the goal of minimizing the sum of active network losses of the distribution network in all time periods within the scheduling time; constructing constraints for the distribution network optimization scheduling model; constructing dynamic equilibrium constraints in the transportation network optimization scheduling model based on the mixed DUE state of the transportation network under an equilibrium path dispatch rate; constructing dynamic equilibrium differential variational inequalities in the transportation network optimization scheduling model; and constructing a dynamic two-layer optimization scheduling model for an electrified transportation-distribution network with the distribution network optimization scheduling model as the upper model and the transportation network scheduling optimization model as the lower model. The present invention controls the temporal and spatial distribution of charging electricity prices and thus controls the comprehensive travel cost of users to achieve orderly charging, optimize traffic flow distribution, effectively reduce distribution network losses, fully utilize the flexibility resources of the dual networks, and smooth out load fluctuations.