83results about How to "Compatible with actual application scenarios" patented technology

The invention relates to a distributed MIMO radar target tracking accuracy joint resource optimization method. The method includes the following steps that: an objective function is derived, a resource allocation optimization model containing three variables, namely, array element, power and bandwidth is constructed; the model is simplified, an array element selection problem is converted into a convex optimization problem from a 0-1 problem through convex relaxation; and the optimization model is decomposed into iterative optimization sub-problems of array element selection and power allocation through using cyclic minimization, the problems are solved through using an SPCA (sequential parametric convex approximation) method until the objective function is not optimized any more, final array element selection and power allocation results are obtained, and a final bandwidth allocation result is calculated, and a target position at a next time point is estimated. According to the distributed MIMO radar target tracking accuracy joint resource optimization method of the invention, the accuracy of the target tracking of an MIMO radar can be effectively improved through resource allocation, and the utilization rate of resources can be also improved; and an appropriate array element subset is selected, and power and bandwidth resources are allocated optimally, and therefore, the accuracy of the target tracking can be further improved under a situation that radar system resources are limited.

The invention provides a Chinese font conversion and generation method based on a small number of target fonts. The method comprises the following steps: (1) preparing an original font stroke data setand an original font-target font data set; (2) selecting training samples with relatively high representativeness by adopting a simple image dimension reduction and clustering algorithm; (3) trainingan end-to-end deep stroke discrete font transformation (DeepSDFT) neural network, and storing network parameters; and (4) automatically generating Chinese characters in the target fonts by using theDeepSDFT neural network to acquire remaining Chinese characters in the target fonts. According to the Chinese font conversion and generation method based on a small number of target fonts, no manual interaction is needed, and no network pre-training is needed; a small number of network parameters are needed; the method can be used for generating various target fonts; interpretability of an embedded feature space is realized; the Chinese font conversion and generation method is applicable to a font design assisted scene and is relatively high in practicability.

The invention discloses a large-scale image sub-block retrieval method based on a deep Hash network and sub-block reordering. The large-scale image sub-block retrieval method comprises the following steps: preparing a deep network weight pre-trained by a large-scale image database and a texture picture library to be trained; extracting picture pairs from the texture picture library, when the two pictures have the same label, determining the two pictures as positive samples, and taking the two pictures with different labels as negative samples; inputting the picture pairs into two same shared weight networks for training in pairs, and setting a loss function to binarize a network output result; quickly obtaining similar pictures by using the network as a Hash code extractor; after subblockreordering is conducted on the similar pictures, quickly mapping subblocks in an original picture and obtaining a large number of similar sub-blocks. According to the method, the calculated amount isgreatly reduced, rapid retrieval of large-scale image sub-blocks is achieved, and the method can be applied to various image enhancement methods.

The embodiment of the invention provides a dissolved oxygen concentration control system, method, device, equipment and storage medium. The system comprises a sewage treatment control platform and multiple water quality testing instruments; the multiple water quality testing instruments are respectively arranged on multiple parameter acquisition positions of a sewage fluid path passage and electrically connected with the sewage treatment control platform; the multiple water quality testing instruments are used for collecting multiple pieces of water quality data of the multiple parameter acquisition positions of the sewage fluid path passage in a first time quantum; and the sewage treatment control platform is used for predicting dissolved oxygen concentration of a biochemical reaction basin in the sewage fluid path passage according to the water quality data in the first time quantum and a pre-trained dissolved oxygen concentration prediction model. Based on the pre-trained dissolvedoxygen concentration prediction model conforming to an actual application scenario, the dissolved oxygen concentration control system can predict the dissolved oxygen concentration in the biochemicalreaction basin more accurately, so that the prediction of the dissolved oxygen concentration is more suitable for the actual application scenario.

The invention belongs to the technical field of a power system and especially relates to a power transmission line wildfire dynamic forecasting and early warning method based on sequential variation. The method is characterized by, to begin with, constructing a wildfire spreading ellipse model at the initial time t0 according to wildfire point initial position F0, wind direction and wind speed and slope information, and calculating wildfire maximum spreading speed Vmax at the current time; calculating the minimum distance D0 between the F0 and a power transmission line L at the time t0 and spreading speed VL0 in the direction, and determining power transmission line wildfire early warning grade at the time t0; determining speed of the fire point F1 in the power transmission line spreading direction at the next time t1; calculating a wildfire spreading model obtained from the time t1 and after a time period delta t; repeating the steps above until obtaining a series of wildfire point positions about time series; and when receiving fire point F0 spreading and ambient environment change information at any time, updating a new wildfire spreading path diagram in time starting from the time. The method can correct wildfire spreading forecasted paths in time, and has the advantages of small calculation amount and clear target and preventing simulation in irrelevant direction.

The invention puts forward a control method and device for information access permission. The method comprises the following steps of creating a user group information table, a dimension permission table and a user group-dimension permission relationship; creating a user information table, and establishing a user-user group relationship and a user-dimension permission relationship; according to a role information table, a role-menu function information table and the created user information table, establishing a user-role relational table, and carrying out user operation permission control according to the user-role relational table; according to the user-user group relationship, the user group-dimension permission relationship and the user-dimension permission relationship, carrying out user data access permission control; and when the user data access permission control is carried out, regulating the user-dimension permission relationship, the user group-dimension permission relationship and the user-user group relationship to regulate the user group and the dimension permission corresponding to the user.

The invention relates to a method and a device for data access permission control, computer equipment and a storage medium. The method comprises the following steps: monitoring a data access instruction triggered based on a service application, wherein the data access instruction carries a user identifier and a data identifier of data to be accessed; determining a role of a target user corresponding to the user identifier in a service application; determining an organization node corresponding to the target user in a preset organization structure tree when determining that the target user hasthe operation authority for the service application according to the role; taking the organization node corresponding to the target user as the organization node of the current level to traverse whether each organization node in the organization structure tree has the access permission to the target data corresponding to the data identifier or not until the organization node of the lowest level isreached; and when one or more low-level organization nodes directly or indirectly connected with the current-level organization node have the access permission to the target data, displaying the target data. By adopting the method, the authority management flexibility can be improved.

The invention discloses a vehicle-mounted terminal locating track fitting method based on a B spline curve. A locating track of a vehicle-mounted terminal is fitted by means of the B spline curve, and therefore the better fitting degree of the actual running track of a vehicle is achieved, a fitted curve has a good smoothness effect, and definite control and filtering characteristics are achieved for errors of local position sampling points. According to the vehicle-mounted terminal locating track fitting method, the three-order B spline curve algorithm is adopted because in a running process of the vehicle, collecting quantity of locating data is large, calculating quantity higher than the three-order B spline curve algorithm is very large, and the smoothness effect on a node is relatively poor when the calculating quantity is lower than the three-order B spline curve algorithm. The three-order B spline curve algorithm has two-order derivative continuous characteristics at the node of a section, is relatively matched with actual running track of the vehicle, and has small calculating quantity, and the vehicle-mounted terminal locating track fitting method is suitable for being applied to engineering.

The invention discloses a geographic region segmentation method based on a road network. The method comprises the steps of extracting a road traffic network from map data of a to-be-segmented geographic region; determining a closed road network corresponding to the road traffic network; performing difference calculation on the to-be-segmented geographic area and the closed road network to obtain an element block set of geographic elements contained in the to-be-segmented geographic area; and matching the element blocks in the element block set with the element attributes contained in the map data to obtain geographic elements corresponding to the element blocks in the element block set. According to the geographic area segmentation method based on the road network, the geographic area is segmented more accurately. Meanwhile, connectivity constraints of road traffic and constraints of actual geographic elements are not damaged, and the geographic area segmentation method better conformsto actual application scenes and is more practical.

The invention discloses an Internet-of-Vehicles V2I mode task unloading method and system. The method comprises the steps that a vehicle user sends a task unloading request to an MEC server matched with a base station through the base station of a road section where the vehicle user is located; the MEC server receives the task unloading request, unloads the task according to the requirement of thetask unloading request, and calculates a task unloading strategy meeting the time delay based on a game theory method; and the MEC server sends the unloaded task to the base station matched with theMEC server, and the base station transmits the unloaded task to the vehicle user according to the task unloading strategy. According to the task unloading method and system, the MEC server uses a calculation result when a task calculation total overhead target is optimal under a calculation delay limitation condition as a task unloading strategy; and the task unloading strategy is issued through the corresponding communication link, so that the complex operation task processing capability of the Internet of Vehicles is improved to a certain extent, the data processing time delay is reduced, the cost expenditure of task unloading is reduced to the maximum extent, and the service performance of the Internet of Vehicles is improved.

The invention discloses a warehousing multi-AGV path planning method based on an improved BP neural network. The method is characterized by comprising the following steps: acquiring point cloud data of a warehousing environment and constructing a grid map; acquiring real-time locations of all AGVs for path planning at the current moment and corresponding target locations thereof, inputting the locations into a trained BP neural network model, and randomly selecting an AGV as a first input to obtain a moving state of the AGV; repeating the above operations for all AGVs whose moving states are not determined at the current moment, till obtaining the moving states of all the AGVs; if all the AGVs for path planning at the current moment reach the target locations, outputting complete paths ofall the AGVs for path planning at the current moment; otherwise, setting next moment as the current moment, and repeating the above operations. The method greatly simplifies the structure of the neural network, improves the accuracy of a learner, simultaneously improves the convergence speed, avoids over-fitting problems, and effectively improves the efficiency of warehousing operations.

The invention relates to an automatic adsorption locking connection mechanism for space truss connection. The automatic adsorption locking connection mechanism comprises a joint piece and a groove piece which are in fit connection with each other, a connecting head at the end part of the joint piece is provided with a petal-shaped protrusion, and the groove piece is of a cylindrical structure; a connecting groove is formed in the groove piece, and a groove in the connecting groove is matched with the protrusion; and magnetic poles used for automatic adsorption are arranged on the position, opposite to the connecting groove, of the connecting head, and magnetic poles used for adsorption and rotating to lock the mechanism are arranged on the connecting head and the side wall of the connecting groove. The automatic adsorption locking connection mechanism is a connecting device which can realize automatic adsorption connection and self locking in a certain range, the structure is reliablyconnected and firmly locked, and the automatic adsorption locking connection mechanism can be reused.

The present invention provides a self-adaptive co-location pattern obtaining method and apparatus considering a spatial instance distance weight. The method comprises the steps of: establishing Thiessen polygons by using instances having the same spatial characteristics as target layers, and establishing a normalization instance dual table on basis of connection relationships of instances of the current target layer and other characteristic instances in a range of influence of the Thiessen polygons; performing distribution parameter calculation on a distance set of the normalization instance dual table, and obtaining normalization distance truncation parameters; establishing a distance feedback function by utilization of the normalization distance truncation parameters, and calculating a distance feedback value of a connection instance couple having any two different characteristics; and obtaining the final co-location pattern based on distance feedback function calculation on the basis of the distance feedback value. The defects of a conventional co-location pattern mining method in aspects of area adaptability, high efficiency and validity are solved.

The invention relates to the technical field of computers, and particularly relates to a service index prediction method and device. The method includes: obtaining a historical service indicator sequence, wherein the historical service index sequence represents service indexes arranged according to a continuous unit time sequence, the unit time is corresponding set fluctuation time when each service index is in fluctuation change, each service index corresponding to the set fluctuation time is in periodic and trend change according to duration time, and the duration time is greater than the set fluctuation time; determining a span difference value between a target moment and a historical reference moment, wherein the historical reference moment is determined from time corresponding to thehistorical service index sequence; and predicting the service index of the target moment based on a cubic exponential smoothing model according to the span difference value, the historical service index sequence and each smoothing parameter determined by training. The periodicity and tendency of the data time sequence can be fully mined based on the cubic exponential smoothing model, and the prediction accuracy is improved.

The invention discloses a weather and time perceived personalized POI sequence recommendation method. The method comprises the following steps of 1, data preprocessing; step 2, POI popularity modeling; step 3, user interest modeling; and step 4, performing of personalized POI sequence recommendation of weather and time perception. The method has the beneficial effects that the method focuses on discussing the influence of weather and time factors on POI sequence recommendation, focuses on considering the weather factors, such as avoiding recommending outdoor POIs to users in rainy days, and focuses on considering the time factors, such as preferentially accessing gymnasiums by the users instead of accessing at night. The personalized POI sequence which is popular and maximizes the interestpreference of the user in a specific weather characteristic and time interval can be recommended to the user, so that the personalized POI sequence better conforms to an actual application scene.

The invention provides a real-time tracking and fast imaging method applied to a camera type millimeter wave human body security instrument, which solves an inversion error problem of the fast Fouriertransform algorithm of the traditional synthetic aperture in a near-field imaging application. The invention proposes a phase compensation method with beam near-field focusing to cooperate with infrared ranging information for near-field focusing and imaging of a synthetic aperture array. The method comprises the following steps: using an infrared camera to collect target distance data, using system calibration information to obtain a relationship between the phase compensation value of the near-field focusing and imaging and the target distance, and realizing the near-field focusing and imaging by performing point-by-point phase compensation of the imaging position according to the calculation result. According to the algorithm of the invention, the real-time tracking and fast imaging ofthe camera type millimeter wave human body security instrument in the near-field can be realized.

The invention discloses a geographic region segmentation method based on a road network. The method comprises the steps of determining a road traffic network in a to-be-segmented geographic region according to region data information of the to-be-segmented geographic region; generating a closed road network corresponding to the road traffic network; performing difference calculation on the to-be-segmented geographic area and the closed road network to obtain an element block set of element blocks corresponding to geographic elements in the to-be-segmented geographic area; and segmenting the element blocks in the element block set according to the region data information. According to the geographic region segmentation method based on the road network, on the basis of considering the connectivity constraint of real road traffic and the constraint of actual geographic elements, finer geographic region segmentation is achieved, and the geographic region segmentation method better conformsto the actual application scene and is more practical.

The invention discloses a method for calculating a regional safety protection capability. The method comprises: safety vulnerability information of all regions is obtained at a predetermined time interval; according to the safety vulnerability information, vulnerability repairing speeds of all regions within all periods of time are calculated; the vulnerability repairing speed of each region is compared with a preset standard value to obtain a comparison result; and on the basis of the comparison results, safety protection capabilities of all regions are determined. In addition, the inventionalso discloses a corresponding computing device.

The invention discloses an address ranking system based on a block chain transaction network and a construction method thereof. The address ranking system comprises a transaction network constructionmodule, a ranking module and a quantitative evaluation module. The transaction network construction module is used for constructing a directed weighted transaction network based on historical data ofblock chain transaction; the ranking module is configured to execute a preset ranking algorithm and carry out important degree scoring on each transaction node in the generated transaction network; and the quantitative evaluation module is used for performing a quantitative assessment algorithm and carrying out quantitative calculation of costs for controlling ranking. With the address ranking system and the construction method thereof, a ranking problem of entities participating in network transaction according to the existing block chain technology is solved; the ranking fairness s improved;a technical defect that the structural value of the user in fund circulation can not be reflected according to the existing block chain ranking technology simply using statistics of the number of times of contract invoking to evaluate the contract quality can be overcome.

The invention discloses an SAR (Synthetic Aperture Radar) constrained energy efficiency maximization multi-user MIMO (Multiple Input Multiple Output) uplink precoding method. In the method, maximization of system energy efficiency is taken as a target, a sending signal covariance matrix of each user in multi-user MIMO uplink transmission is designed by utilizing statistical channel state information, and meanwhile, the constraint of the maximum sending power and the constraint of an electromagnetic wave absorption ratio SAR are met. The energy efficiency maximization algorithm design mainly utilizes a deterministic equivalence principle, Dinkelbach transformation and an iterative water injection algorithm to obtain an optimal sending signal covariance matrix. When the channel state information is changed, the user side dynamically implements the energy efficiency maximization multi-user MIMO uplink transmission method meeting the SAR constraint. According to the method, on the premiseof meeting SAR constraints, the energy efficiency value of multi-user MIMO uplink transmission can be effectively improved, and the implementation complexity of wireless transmission can be remarkablyreduced.

The invention discloses an ultra-low-altitude flight area classification setting method for an unmanned aerial vehicle, and the method comprises the following steps: carrying out the rasterization ofthe ultra-low-altitude flight area, and carrying out the unique identification of each grid; establishing unmanned aerial vehicle ultra-low-altitude flight area classification setting evaluation indexes and index quantitative calculation models to obtain n evaluation indexes and c risk level value intervals corresponding to the n evaluation indexes respectively; utilizing a G1 method to obtain each index weight; determining an adjacent grid addition characteristic value according to the risk evaluation index value interval corresponding to each grid; calculating the grade relative membership degree of the indexes corresponding to the grids; calculating to obtain an unmanned aerial vehicle flight limitation degree level characteristic value corresponding to the grid; and establishing an unmanned aerial vehicle flight limit degree level characteristic value threshold value model, comparing the unmanned aerial vehicle flight limit degree level characteristic value corresponding to the grid with the unmanned aerial vehicle flight limit degree level characteristic value threshold value, and determining an unmanned aerial vehicle ultra-low-altitude flight area classification division level number corresponding to the grid. The unmanned aerial vehicle ultra-low-altitude flight area classification setting method based on quantitative analysis is established by comprehensively considering three factors of the unmanned aerial vehicle over-the-ground risk, the over-the-air risk and the operation safety of the unmanned aerial vehicle.

The invention discloses an edge node calculation unloading method with performance and demand matching capability. The method comprises the following steps: establishing a cooperative computing network system with a plurality of mobile edge node devices; constructing and initializing a node performance estimation model and a task demand estimation model, and initializing required operation variables; carrying out online selection of unloading targets by using a reinforcement learning method based on an adversarial dobby machine; unloading the task to the selected target, and observing the actual unloading delay; updating a node performance estimation model by using the actual delay of the unloading; the central equipment using the updated node performance estimation model as a behavior basis for next unloading; whether a new unloading request exists or not being judged, if yes, online selection of the unloading target is carried out again, and if not, the new unloading request continues to be waited. According to the method, the traditional single unloading delay is deconstructed into independent double-variable unloading delay to be considered, and the method is better in line with the actual application scene of the Internet of Things equipment.

The invention discloses an RIS-assisted multi-user downlink robust wireless transmission method based on partial CSI. The RIS-assisted multi-user downlink wireless transmission method comprises the following steps: firstly, constructing an RIS-assisted multi-user downlink wireless transmission system comprising a multi-antenna base station, an RIS consisting of a plurality of passive reflection units and a plurality of single-antenna users; secondly, aiming at an RIS-assisted multi-user downlink wireless transmission system, a plurality of users and a base station respectively transmit pilot signals to the RIS, and the RIS respectively estimates CSI with errors from the RIS to the base station and the users; and finally, maximizing the large system approximation sum rate of multiple users as an optimization target, designing a beam matrix and an RIS phase matrix by means of the estimated CSI, and carrying out signal transmission. According to the method, non-ideal CSI is considered, minimization of transmission power is taken as an optimization target, an actual application scene is better met, phase shift and beam vectors are alternately optimized, the method is more robust, the effect of saving energy consumption is achieved, and the lowest requirement of user safety communication is guaranteed.

The invention relates to an RGBD image joint recovery method based on a double-flow network. The method comprises the steps of: S1, obtaining an RGBD image database used for training and testing; S2,dividing the RGBD image database into a training data set and a test data set, and preprocessing RGBD images of the RGBD image database; S3, training a double-flow convolutional network model according to the training data set, and storing the trained network parameters; and S4, inputting the test data set into the double-flow convolutional network model for joint recovery, and testing the recovery degree. According to the method, the degraded RGB image and the corresponding depth image can be restored at the same time, and the method conforms to the actual application scene.

The invention provides a server disk performance testing system and method based on Hadoop. The whole environment needs a master server node, a slave server node, and a slave server node as a server to be tested. The two servers need to be in a network segment to facilitate interconnection. The test environment is simple. If there are more servers to be tested, multiple slave server nodes can be set to test synchronously. The method comprises the following steps: S1, configuring an environment of a server disk performance testing system; S2, performing performance testing on the server disk; S3, processing and analyzing test data and log. Hadoop, an open source framework, is adopted to verify the performance of the disk on the server, and verify the performance of the selected disk on thedistributed storage application scenario by approaching the actual client application scenario. Through the tools of an Hadoop program, files with different sizes are simulated, the performance of disk in various scenarios is verified, and the theoretical data of disk is compared, and evaluation and suggestions are provided.

The invention discloses a data distribution type incremental learning method, system and device and a storage medium. The method comprises the following steps: determining the type of each incremental learning stage, and building a data set of each data distribution node; obtaining each data distribution point model; forming a shared data set; obtaining model parameters of each data distribution node; performing weighted aggregation on the model parameters of the data distribution nodes to obtain a preliminary global sharing model; integrating predicted output logit values obtained by calculating the M data distribution node models on the shared data set to obtain an integrated output logit value, and learning the integrated output logit value on the shared data set by the preliminary global shared model to obtain model parameters of the global shared model; the model parameters of the global sharing model are issued to all the data distribution nodes, the global sharing model on all the data distribution nodes is updated, and the method, the system, the equipment and the storage medium can effectively improve the learning ability of the model.

The invention discloses an electrical interconnection integrated energy system optimization planning method and system. An integrated energy system optimization planning model in which a power system and a natural gas system are combined is established, and the model is a two-stage three-level defense-attack-defense robust optimization planning model, and meanwhile, include extension plans of power lines and natural gas pipelines; in the first layer, a system planning manager needs to reasonably carry out joint planning on a power system and a natural gas system to minimize the maximum load loss caused by any possible extreme event; in the second layer, an attacker attacks the fragile part in the power system and the natural gas system to cause the maximum load loss; and on the third layer, a system operation manager can optimize the internal operation state of the system for an extreme scene to achieve the minimum load loss in a damaged state. The invention lays a solid foundation for building an efficient, continuous and safe modern energy supply system.

The invention provides a molecular docking method and system based on migration learning. The method comprises: acquiring 3-dimensional space coordinates, van der Waals radius and atom types of a plurality of docking samples; dividing the channel number according to the atomic type of the docking samples, and calculating corresponding values on each grid point according to the 3-dimensional spacecoordinates and the van der Waals radius to obtain multi-channel 3-dimensional grid data; inputting the multi-channel 3-dimensional grid data into the network architecture, outputting the scores of each docking samples and classifying the docking samples according to the scores, so as to train the docking model, wherein for the first trained docking model used for different families of migration learning, different docking models are trained for different families of different docking samples. The invention can automatically train a docking model according to the docking samples, so that the docking model can accurately complete the molecular docking and conform to the practical application scenarios of different families of samples, and improve the accuracy of molecular docking.

The invention relates to a broadband micro-power multi-network comprehensive test system for an electricity utilization information acquisition system, which belongs to the field of electricity utilization information acquisition of power consumers and comprises a system configuration module, a test case module, a data analysis module, a test report module, an adaptation module, an adjustable attenuator, a signal isolation box, a simulation ammeter module and a base. An external multi-network environment is built through the concentrator, the electric energy meter, the adjustable attenuator, the signal isolation box and the like, so that multi-level multi-network can be conveniently realized, and an actual application scene is met. In order to realize full coverage of a multi-network test, a plurality of communication modules are placed through a base, and an electric energy meter which is in one-to-one correspondence with the base communication module is simulated at a PC end, so that a tester can easily control a tested object, and the simulated electric energy meter can make up some processes that a real electric meter cannot be tested in a laboratory environment.