158results about How to "Solve bottlenecks" patented technology

The invention relates to the field of transformer parameter regulation and control, in particular to a parameter optimization method and device for a coupling type transformer. The method comprises the following steps: acquiring multi-dimensional real-time working monitoring parameters and real-time working environment state parameters of the coupling type transformer; performing nonlinear dynamic analysis and global state sensing modeling on the multi-dimensional real-time working monitoring parameters so as to construct a transformer global state sensing model; performing multi-parameter combination definition on the multi-dimensional real-time working monitoring parameters, performing dynamic operation simulation based on a transformer global state sensing model, and collecting operation simulation response data of a plurality of schemes; performing multi-target dynamic balance analysis on the operation simulation response data of the plurality of schemes to obtain an optimal transformer working parameter combination; and carrying out environment dynamic parameter change analysis on the real-time working environment state parameters, and constructing a real-time environment condition distribution field. According to the invention, efficient and self-adaptive adjustment transformer parameter optimization is realized.

The invention discloses a Transform remote sensing semantic segmentation method based on an adaptive dynamic attention mechanism, and relates to the technical field of the adaptive dynamic attention mechanism, and the method comprises the following steps: obtaining a remote sensing image data set, and carrying out the data marking and data enhancement; adjusting the remote sensing image into a model input size and a pixel value unified specification, and generating a standardized data set; feature embedding is generated based on a preset embedding module, and position coding is added to enhance position information for decoding; correcting the decoding process by using a self-attention and cross attention module; combining different scale features with continuously iterated updated adaptive weights to obtain a fusion feature value; constructing a multi-task joint loss function based on the to-be-solved problem; refining the fusion feature value through a segmentation module, and generating predicted values of different sub-tasks; according to the method, the problems of information loss and inconsistency possibly occurring during feature fusion in a multi-scale task in a traditional method are solved.

The invention belongs to but not limited to the technical field of electrode materials, and particularly relates to a high-tap-density lithium nickel manganese oxide positive electrode material, a preparation method and a lithium ion battery, the chemical general formula of the high-tap-density lithium nickel manganese oxide positive electrode material is LiNi < 0.5-x > Mn < 1.5-y > M < x > N < y > O < 4 >, 0.1, 0 < = y < lt >; 0.1, M and N are doping elements and comprise one or more of Mg, Co, Al, Zr, W, Sb, Ta and Ti; the material is micron-sized spherical secondary particles formed by stacking and aggregating nano-sized octahedral primary particles, and is uniformly dispersed without agglomeration; the spherical secondary particles are formed by mixing spherical secondary particles with large and small particle size distribution, and the average particle size range of the large particles is 3lt; d < 50lt >; the average particle size range of the small particles is 1lt; d < 50lt >; 2 [mu] m.

The invention relates to the technical field of software testing, and discloses a computer software performance testing method which comprises the following steps: S1, determining performance testing requirements: determining key performance indexes according to application requirements, and designing a layered load model; s2, constructing a dynamic test environment: realizing elastic adjustment of resources by utilizing a virtualization technology; s3, distributed performance testing: executing concurrent testing at a plurality of nodes, and simulating access of different users; s4, transverse extension testing: adjusting a virtual machine or a container, and evaluating the system extension capability; s5, collecting and analyzing performance data: analyzing and monitoring the performance data through a time sequence, and identifying a bottleneck; and S6, generating an optimization suggestion and a regression test: automatically generating the optimization suggestion, and verifying the effect through the regression test. Through combination of a virtualization technology, Kubernetes and OpenStack and cooperation with a lateral extension test, the stability and the extension capability of the system under a high concurrency condition are verified, and the overall operation efficiency is improved.

The invention discloses a brain-like edge calculation-driven multi-mode intelligent security method and system, and aims to improve the security capability in a complex dynamic scene. According to the method, multi-modal data of videos, audios, sensors and the like are collected, preprocessing is carried out by utilizing a brain-like edge computing gateway, dynamic behavior characteristics are extracted, secondary confirmation is carried out through a lightweight visual large model, and then formal early warning is generated. And the system executes intelligent linkage operation according to the early warning information risk level, and records false alarm information for model optimization. According to the invention, brain-like calculation and multi-modal fusion are adopted to improve the recognition precision, edge calculation reduces delay, an intelligent closed loop realizes differentiated treatment, a low-power design and a modular architecture reduce the cost, and the system has a self-learning capability. The defects of an existing security and protection system in the aspects of dynamic behavior recognition, response speed, disposal intelligence, cost control and the like are effectively overcome. The system is suitable for scenes such as a campus and a community, and provides an accurate, efficient and intelligent security solution.

The invention discloses a computer network security transaction optimization system, and belongs to the technical field of network transaction security. The system comprises an identity verification authorization module, a transaction request processing module, a security monitoring protection module, a transaction intelligent optimization module, a transaction auditing management module and a transaction report analysis module, protection is enhanced by analyzing a user behavior mode, additional verification is triggered when abnormality is detected, the risks of identity false use and fraud are reduced, and the safety of the transaction is improved. Network traffic is monitored in real time, potential attack behaviors are identified and blocked, through transaction request checking and dynamic request scheduling strict auditing, node loads are monitored in real time, transaction requests are intelligently scheduled, system overload or bottleneck is avoided, standby nodes are automatically started, stable operation under the high concurrency condition is guaranteed, and the system reliability is improved. The method enhances the adaptation to the increase of the transaction flow and the change of the network environment, improves the security, efficiency and transparency of the transaction through the transaction audit management and report analysis, and guarantees the stability and compliance of the financial transaction environment.

The invention belongs to the technical field of stratum recognition, and particularly relates to a shield parameter geological recognition calculation method and system based on an unsupervised clustering algorithm, and the method comprises the steps: selecting a plurality of groups of data containing factors affecting stratum category recognition; pCA converts a group of complex related variables into fewer and irrelevant variables, and the new variables are called as main components and can capture most of information in original data; determining the clustering number of K-means + + by adopting an elbow method; determining the clustering number of a hierarchical clustering method by adopting a hierarchical tree; k-means + + and hierarchical clustering are respectively used for clustering the data, and evaluation indexes such as identification precision or contour coefficients of the data and the data are calculated, so that potential modes in the data are analyzed and mined more comprehensively. The invention aims to deeply explore the internal relevance between shield tunnel parameters and geological types, provides an innovative and robust solution for real-time detection of geological features in shield construction by overcoming the limitation of the prior art, and has wide application potential.

The invention discloses a distributed collaborative modularized rapid detection method and device for a box-type substation, and the method comprises the following steps: enabling each distributed module to judge whether the distributed module has a fault possibility or not based on collected operation data; the distributed module with the fault possibility initiates an abnormal signaling; after the other distributed modules receive the abnormal signaling, whether correction characteristic parameters of the other distributed modules are filled in corresponding positions in the abnormal signaling or not is confirmed, and the abnormal signaling is sequentially transmitted among all the other distributed modules; and summarizing all the obtained correction characteristic parameters, calculating a posterior fault probability by the distributed module which initiates the abnormal signaling, and judging that a fault exists when the posterior fault probability exceeds a set threshold value. According to the distributed architecture, the bottleneck of a centralized system is avoided, the modules are processed in parallel, and the detection efficiency is improved. Through the feature correlation model, interference of redundant information is avoided, and the accuracy of fault judgment is improved.

The invention is suitable for the technical field of data transmission, and provides an illumination lamp data transmission control method, which comprises the following steps of: acquiring data, state data and physiological health data by an illumination lamp; different priorities are set for each type of data, and different data are processed by adopting corresponding compression strategies; a multi-hop transmission ad hoc network is established between the lighting lamp and other nearby lighting lamps, and a network topology structure is dynamically adjusted; the lighting lamp evaluates an optimal sequence of various transmission paths transmitted to the monitoring center based on real-time information in the ad hoc network, and sets corresponding transmission paths and transmission strategies for different data according to priorities; the lighting lamp transmits the data with different priorities to the monitoring center hop by hop through a selected transmission path according to a set transmission strategy; and the monitoring center receives the data of the lighting lamps and carries out corresponding processing according to the priority of the data. According to the invention, the problems of stability, real-time performance and priority management of data transmission in an existing complex mine environment are solved.

The invention relates to the field of intelligent traffic management, and discloses an efficient road obstacle clearance rescue worker automatic scheduling method comprising the following steps: (a) deploying at least one message queue to cache rescue related data; (b) storing and querying rescue tasks and resource state data through a distributed database system; (c) collecting data from the rescue vehicle and the accident site in real time, and performing preprocessing; (d) asynchronously processing the collected data through a message queue according to the priority and the task type; (e) dynamically scheduling rescue resources by using a multi-factor decision algorithm; the message queue adopts a Kafka cluster and is configured with at least three partitions and at least two copies so as to ensure that a single server or partition fails. Through combination of a multi-factor decision scheduling algorithm and real-time traffic flow prediction, the system can dynamically adjust the scheduling route of the rescue vehicle, and the rescue vehicle is prevented from entering a congested road section, so that the time for arriving at an accident scene is remarkably shortened, and the overall rescue efficiency is improved.

The invention discloses a self-supervised clustering method for massive single cell data. The method comprises the following steps: 1) downloading single cell RNA sequencing data from a disclosed gene expression database; 2) preprocessing the single-cell RNA-seq data, wherein the preprocessing comprises filtering, standardization and feature selection; (3) carrying out preliminary clustering on the preprocessed single cell RNA sequencing data by utilizing a rapid Louvain algorithm; 4) screening representative cell points from the preliminary clustering result by using an index of clustering representativeness; 5), after representative cell points are screened out from each preliminary cluster, a Transform model is used, and a final clustering result is obtained.According to the method, the rapid Louvain algorithm and the Transform neural network are combined, the efficiency and accuracy of clustering analysis are remarkably improved, and an efficient and accurate solution is provided for analysis of mass single cell data.

The embodiment of the invention discloses a distributed self-increasing sequence synchronization method and device, equipment and a storage medium. The method comprises the steps that in a distributed database cluster, a master node and a plurality of slave nodes are arranged, and the master node is used for maintaining a global self-increasing sequence value; the slave node requests to distribute a local self-increasing sequence to the master node according to a preset cache parameter, and the master node calculates an initial value of the local self-increasing sequence according to the global self-increasing sequence value and distributes a preset number of continuous self-increasing sequence values to the slave node; the slave node caches the local auto-increment sequence to the local; and when the local self-increasing sequence cached by the slave node is used up, sending a new allocation request to the master node, and reallocating a new local self-increasing sequence by the master node according to the current global self-increasing sequence value. According to the method, the master node and the slave nodes are arranged in the distributed database cluster environment, and the global self-increasing sequence value is uniformly managed by the master node, so that reliable generation of the self-increasing sequence in the distributed environment is realized.

The invention belongs to the technical field of federal learning, and discloses an intelligent model training method and system for distributed equipment collaborative optimization and a storage medium, and the algorithm comprises the steps that a client receives an original model sent by a server, and completes an initialization task of the model; the client carries out local training on the received initialization model based on the local data set to obtain a local training model, all basic parameters of local training are obtained through calculation, and a memory-free SR1 matrix is generated; the server randomly extracts part of clients, and the selected clients upload local model training parameters of the clients; updating an ADMM direction, and updating an internal problem; solving a variable-scale near-end mapping problem to obtain a near-end direction, and solving an external problem; and updating the global model according to the obtained near-end direction until the model training is completed. Under the condition that statistical heterogeneity exists, compared with an existing first-order composite optimization algorithm, the algorithm has the advantages that the convergence speed is higher, the convergence condition of the algorithm can be met more quickly, and progressive convergence is achieved.

The invention provides a satellite-borne router initialization method. The satellite-borne router initialization method comprises the following steps that a comprehensive electronic system sends a power-on instruction to a satellite-borne router; an initialization module of the satellite-borne router starts to read a preset initialization instruction set, and preparation work before initialization is carried out; after the time T1, the satellite-borne router completes the early-stage preparation of initialization and enters an init-bus state; after the time T2, a hardware module of the satellite-borne router completes a power-up process; after the time T3, the spaceborne router obtains an IP address through a DHCP protocol; performing system inspection and self-inspection; and completing initialization of the satellite-borne router. The invention provides an initialization method of a satellite-borne router. The method improves the starting stability and efficiency of the satellite-borne router.

The invention belongs to the field of indoor environment health research, and particularly relates to a method for effectively representing a volatile organic compound (VOCs) release mechanism of an indoor material under long-term physical and chemical action. According to the method, accurate and effective characterization of the long-term release characteristic of the material is realized by simplifying a complex physical and chemical coupling problem and combining an aging test and a parameter determination experiment. The method comprises the following steps: designing a natural aging experiment to simulate an actual long-term release process of a material; extracting material samples with different aging times, and determining corresponding VOCs key parameters by utilizing an environment chamber experiment; and fitting the change of the key parameters along with time by using a single exponential function so as to characterize the long-term release characteristic of the VOCs in the material. The method is suitable for material-VOCs combined release research under different temperature and humidity conditions, and results can be used for long-term concentration prediction of indoor VOCs in a dynamic environment and further applied to reasonable material selection in building design, ventilation design in specific space and time, health risk assessment and the like.

The invention discloses a data acquisition and automatic control system for a carbon filament collecting machine, which adopts a modular design and comprises a main control module for connecting and controlling other modules, and a plurality of functional modules such as a display and operation interface module, a real-time monitoring module, a fault detection and early warning module, a shredding control module, a label printing module and a multi-device adaptation module. The system is realized on the basis of an MVVM (Model-View-ViewModel) architecture of a WPF (Wireless Power Function), and has good expansibility and maintainability. The system uses the SQL Server as a database, and the EFCore performs data operation, so that data access is more efficient and flexible. Meanwhile, the system supports parallel operation of a plurality of wire collecting machines, each wire collecting machine is managed through an independent thread, concurrence conflicts are avoided, PLC data can be obtained in real time, wire collecting machine equipment can be dynamically managed, and a proper label template is automatically selected for printing according to batch information.

The invention belongs to the field of human health exposure research, and particularly relates to an efficient method for measuring adsorption and release parameters of volatile organic compounds (VOCs) of skin or artificial skin. The method comprises the following steps: designing a skin loader for bearing experimental skin and keeping the experimental skin in a hydration state in an experimental process; a testing system suitable for skin VOCs adsorption and release is developed; establishing a physical model for describing a skin VOCs release process; the skin to be measured is placed in the constant-temperature environment chamber, sufficient VOCs gas is injected to enable the skin to be adsorbed to a balanced state, then pure air is introduced to promote the skin to release VOCs, and meanwhile the gas-phase VOCs concentration in the chamber is measured. And carrying out nonlinear fitting on the concentration data by combining particle swarm optimization and an ant colony algorithm, and calculating diffusion coefficients (Dm) and distribution coefficients (K) of different VOCs in the skin. The testing system is low in cost, easy and convenient to operate, high in calculation method precision, good in expansibility, suitable for testing of other materials and capable of being widely applied to laboratories and engineering application.

The invention relates to the technical field of physiological signal processing, and discloses a millimeter wave radar multi-dimensional physiological feature decoupling heart failure early warning system and method, and the method comprises the steps: configuring a millimeter wave radar to transmit a dual-beam frequency modulation continuous wave signal, collecting human body reflection echoes, and generating a multi-channel original digital signal matrix; generating a chest region and acra region phase signal, an overall reflection signal amplitude sequence and an overall phase drift signal after spatial isolation enhancement through distance-Doppler combined processing; decoupling to obtain a beat-by-beat heartbeat interval sequence, a pulse wave transmission time sequence and a daily net body fluid variation sequence; obtaining a rule engine risk level and a machine learning early warning sign, and determining a final risk assessment level through fusion decision; synchronous high-precision collection and decoupling of non-contact multi-dimensional physiological parameters are achieved, and accuracy and specificity of heart failure decompensation early warning are improved through double-layer model fusion.

The embodiment of the invention provides a mass data processing method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a plurality of pieces of first data information to be processed and an identification number corresponding to each piece of first data information; according to a preset queue state in a blocking queue, sequentially adding identification numbers corresponding to the first data information into the blocking queue, and updating the queue state; according to the queue sequence of the blocking queues, pushing identification numbers to message queues of preset message middleware in sequence, and updating the blocking queues and queue states; converting the first data information corresponding to the identification number into second data information according to the identification number in the message queue; and pushing the plurality of pieces of second data information and the corresponding identification numbers to a task thread pool for task allocation and processing. According to the embodiment of the invention, task division can be coordinated through the message middleware, parallel processing of mass data is realized, the processing efficiency is improved, and the method is simple to use and convenient to expand.

The invention relates to the technical field of carbon management, in particular to a carbon metering terminal based on edge intelligent collaboration, and the terminal comprises a model partition updating module which constructs a carbon metering model based on carbon emission data collected in real time according to the carbon emission data, and generates carbon metering information after partition optimization by adjusting and dividing the carbon metering model. According to the invention, through real-time acquisition and partition optimization of the carbon emission data, in combination with an extraction mode of time sequence features and statistical features, the precision and efficiency of carbon emission prediction are significantly enhanced. In a data processing process, abnormal data and repetitive data are effectively eliminated by establishing a filtering threshold range, so that the data quality is improved, and the interference of low-efficiency data on subsequent analysis is avoided. In collaborative optimization of the edge nodes, communication delay and resource waste among the nodes are reduced by means of dynamic weight adjustment and path optimization, and the transmission efficiency and the collaborative effect are improved.

The invention discloses a data read-write method and device in a storage system, electronic equipment and a computer readable storage medium, the storage system comprises a client, a main node and a plurality of storage nodes, the client and each storage node are connected to the main node, and the method is applied to the main node. Comprising the following steps: determining target storage nodes according to read-write control information sent by a client, and generating data read-write requests corresponding to the target storage nodes; if the data reading requests are data reading requests, sending the data reading requests to corresponding target storage nodes, so that the target storage nodes send first target data corresponding to the data reading requests to the client; if the data writing requests are data writing requests, sending the data writing requests to corresponding target storage nodes, so that the target storage nodes obtain second target data corresponding to the data writing requests from the client and store the second target data; and receiving a read-write completion message fed back by each target storage node and feeding back the read-write completion message to the client. According to the scheme, the data read-write efficiency in the storage system can be effectively improved.

The invention relates to the technical field of communication fusion, in particular to a multi-link communication fusion scheduling system, which comprises a command terminal, a fusion scheduling module and a communication terminal, and is characterized in that the fusion scheduling module receives communication data of the communication terminal, carries out fusion processing on the received communication data and sends the communication data to the command terminal; a communication link with the optimal current coverage signal in ad hoc network communication, cellular mobile communication and satellite communication is used for sending the communication data to the command terminal, receiving the communication data of the command terminal and sending the communication data to various communication terminals respectively; through the fusion scheduling module, a multi-network fusion communication means of ad hoc network communication, cellular mobile communication, satellite communication and high-frequency radio communication is realized, and the efficient, rapid and convenient integrated combat application effect of mobile combat is enhanced. And the front-end early warning, defense, countering and striking capabilities can be effectively improved, and the efficiency that central commanders can simultaneously issue instructions to different terminals and personnel in multiple parties and control the commanding efficiency can be effectively improved.

The invention discloses a forestry remote sensing data lightweight method based on submerged space optimization, and belongs to the technical field of image processing, and the method comprises the steps: obtaining a to-be-compressed forestry remote sensing image data set D1; constructing a data lightweight network, wherein the data lightweight network comprises a variational auto-encoder, a clustering unit, an image classification network, a multilayer feature extraction network and a discriminator; training a variational auto-encoder and an image classification network by using the D1; generating a clustering data set D2 based on the encoder and the clustering unit; and training the data lightweight network to update the submerged space variable in the D2, and reconstructing the data by using a decoder. Through data set distillation improvement based on the variational auto-encoder and element gradient optimization, a compressed data set with high precision and representativeness is generated, the bottleneck problem of forestry remote sensing data set storage and processing is effectively solved, and the data storage amount can be reduced while the model training precision is kept. The method is suitable for various practical application scenes such as forestry resource monitoring, forest health assessment and disaster response.

The automatic feeding device for belt pulley production comprises a transfer trolley and a feeding table, a belt pulley bracket is arranged on the transfer trolley, a material distributing mechanism and a transferring mechanism are arranged on the feeding table, the material distributing mechanism supports a belt pulley on the belt pulley bracket, and the belt pulley is grabbed and transferred to machining equipment through the transferring mechanism. According to the belt pulley transferring and feeding device, automation of the belt pulley transferring and feeding process is achieved, manual operation is not needed any more, waiting time and errors of manual operation are reduced, continuity and high efficiency of the production process are guaranteed, the bottleneck problem of transferring and feeding operation between procedures is effectively solved, and remarkable production benefit improvement is brought to enterprises.

The invention discloses an intelligent capacity expansion SDN gateway system for a cloud computer, belongs to the technical field of cloud computing, and aims to solve the technical problem of how to realize capacity expansion of an SDN gateway in a cloud computer environment so as to deal with network traffic congestion. Comprising an index acquisition monitoring module used for acquiring key performance indexes from an SDN gateway, the key performance indexes comprising a CPU usage rate, a memory usage rate, a disk read-write rate and a real-time network flow; the intelligent prediction module is used for training and optimizing the network traffic prediction model based on the historical key performance indexes and the user use habit data, and predicting the traffic trend of the SDN gateway in a future predetermined time period based on the trained and optimized network traffic prediction model; and the intelligent decision-making module is used for carrying out expansion decision-making based on the key performance index, the flow trend of the SDN gateway in the future predetermined time period and the resource utilization rate.

The utility model provides a mirror surface roller roundness detection base belongs to roundness detection technical field. A mirror surface roller roundness detection base, including at least two three -dimensional parallel settings recess fixed block, recess fixed block side upper vertical through the at least three positioning slot, two the positioning slot is supported in the rotation connection with the rolling bearing through the limiting shaft. The utility model is opposite to the roller body of different length, need not to adjust the position of tailstock even to replace the whole tailstock component, only need to utilize the on -off type magnetic attraction base quick adjustment its span can. Its adjustment process is simple, can satisfy modern production line fast rhythm, the detection demand of many varieties.

This invention relates to transition layer materials and welding processes for welding dissimilar steel and aluminum materials, belonging to the field of welding technology. The transition layer material for welding dissimilar steel and aluminum materials of this invention comprises a copper layer, a copper alloy layer, and a zinc alloy layer arranged sequentially; the copper layer is adjacent to the steel base material, and the zinc alloy layer is adjacent to the aluminum base material; the copper content in the copper alloy layer is 62.0%-64.0%; the zinc content in the zinc alloy layer is 89.0%-91.0%; the mass ratio of the copper layer, copper alloy layer, and zinc alloy layer is (0.9-1.1):(2.3-2.7):(1.8-2.2). By sequentially setting the copper layer, copper alloy layer, and zinc alloy layer between the steel base material and the aluminum base material, a gradual transition from the steel base material to the aluminum base material is achieved, effectively avoiding the formation of brittle Fe-Al intermetallic compounds during direct welding, and enhancing the mechanical properties and durability of the welded joint.