218results about How to "Reduce time complexity" patented technology

The invention belongs to the technical field of power grid main equipment diagnosis, and provides a power grid main equipment fault diagnosis method and system based on voiceprint cloud edge collaboration, which adopts a new cloud edge collaboration method and divides the analysis of voiceprint monitoring data into two stages. According to the method, the exception detection model is deployed on the edge side, exception analysis of mass voiceprint data is achieved, concurrent requests of a large number of concurrent signals at the cloud side are avoided, and the timeliness of voiceprint monitoring and analysis is improved. In the cloud, a multi-modal defect identification method fusing voiceprint, visible light, infrared and other multi-source data is deployed, secondary classification of sound abnormal signals is realized, and the reliability of equipment operation, maintenance and management is improved.

The invention discloses a power prediction method considering photovoltaic power clustering, which comprises the following steps of: firstly, acquiring historical data of a photovoltaic power station, decomposing a photovoltaic power time sequence by using an STL time sequence decomposition method, performing DBSCAN clustering on the decomposed photovoltaic power data, identifying a low-power cluster and a high-power cluster, and predicting the power of the photovoltaic power station according to the low-power cluster and the high-power cluster; establishing a photovoltaic power prediction model based on a time convolutional network and Times Net for low power and high power respectively; secondly, a badger algorithm is improved, Chebyshev chaotic mapping is adopted to improve initialization of a badger population, and a Gaussian mutation strategy and a lens imaging reverse learning strategy are introduced to obtain MHBA; and optimizing hyper-parameters of the TCN photovoltaic low-power prediction model and the TimesNet photovoltaic high-power prediction model by using an improved badger algorithm, and predicting photovoltaic power by using the optimized models to obtain a final prediction value. Compared with the prior art, the precision and effectiveness of the photovoltaic power prediction model can be improved.

The invention relates to the field of image clustering analysis, in particular to an image clustering method and device based on a particle multi-kernel k-mean value and electronic equipment. The method comprises the following steps: acquiring a discrete data point set of a plurality of to-be-clustered images; performing granular ball division on the image data to obtain an image granular ball set; calculating to obtain an image pellet center, an image pellet distance matrix and an average distance between image pellets; calculating to obtain an image pellet kernel matrix; according to the image pellet kernel matrix and a preset clustering division matrix, constructing a multi-kernel k-means clustering objective function based on pellets; calculating to obtain a corresponding target clustering division matrix when the target function value is minimum; discretizing the target clustering division matrix according to a multi-kernel k-mean algorithm to obtain a clustering result of image pellets; and distributing the clustering result of the image pellets to all discrete data points contained in the image pellets to obtain the clustering result of the image data. According to the invention, the efficiency and accuracy of image clustering are improved.

The invention discloses an optical fiber terahertz communication system optimization method based on artificial intelligence, and belongs to the technical field of communication. According to the method, four-dimensional constellation shaping is constructed by utilizing additional dimensions provided by two independent sidebands, and an end-to-end optimization framework is divided into a training stage and a deployment stage; in the training stage, firstly, an equalizer based on a pruning neural network and an equalized channel model based on a mixed density network are trained in sequence through receiving and transmitting of a random training sequence; then connecting a constellation shaper at a transmitting end with a demapper at a receiving end through an equalized channel model with fixed parameters, thereby considering full-link residual distortion and noise after equalization, and carrying out back propagation gradient to learn a globally optimal four-dimensional joint geometric probability constellation diagram to further approach the Shannon limit; and in the deployment stage, transmission of the bit stream is realized through coordination of the four-dimensional geometric shaping constellation table and the probability distribution obtained in the training stage and an equalizer at the receiving end and a demapper.

The invention relates to a hyperspectral image classification method based on a meta-pseudo-label multi-scale image convolution network, and the method comprises the steps: building a meta-pseudo-label multi-scale feature search image convolution network (MPL-MFSGCN), dividing the HSI into superpixel images of different scales through employing an SLIC superpixel segmentation technology, and converting the superpixel images into image structures; secondly, constructing a multi-scale feature search graph convolutional network MFSGCN architecture, optimizing a superpixel feature weight combination by using a neural network search algorithm, and enhancing the discrimination of feature expression; a semi-supervised teacher-student model sharing the same graph convolutional network is adopted to generate meta-pseudo tags, and a student network is finely adjusted to improve the robustness of the network; the key features can be automatically captured under the small sample condition, the classification precision higher than that of an advanced method is obtained compared with a common HSI data set, and the classification accuracy is effectively improved.

The invention discloses a reliability parameter assisted low-complexity polarization code rate matching method, and belongs to the technical field of wireless communication. Aiming at the problem that a proper punching site is difficult to select during rate matching due to a recursive structure of a polar code, the invention provides a polar code punching site selection algorithm based on a reliability parameter. An information bit reliability parameter is defined to calculate the reliability of candidate punching bits, and the optimal punching position is selected by traversing and solving the minimum value of the maximum reliability parameter. In order to solve the problem of high complexity of the existing polar code rate matching method, an interleaving algorithm and a pruning serial offset list decoding algorithm based on rate matching parameter mapping are provided, secondary sorting calculation of an interleaver is avoided, and a calculation unit updating strategy of a decoder is simplified. The polarization code rate matching method provided by the invention not only can improve the rate matching coding gain, but also can effectively reduce the calculation complexity.

The invention discloses a federated learning contribution degree evaluation method and device, electronic equipment, a medium and a product, and relates to the technical field of federated learning, and the federated learning contribution degree evaluation method comprises the steps: obtaining an initial global model, and sending the initial global model to each client in a preset client set; performing federal iteration training on the initial global model based on a received local model uploaded by each client in a preset client set, and determining an overall global model; performing federal iteration training on the initial global model based on received local models uploaded by clients except the to-be-evaluated client in a preset client set, and determining partial global models; and determining the contribution degree of the to-be-evaluated client according to a performance difference value between the overall global model and the partial global model. According to the federal learning contribution degree evaluation method provided by the invention, the calculation complexity of the contribution degree is reduced, and the time consumption and the calculation resource demand are reduced.

The invention discloses a method for solving a task-platform optimization configuration problem based on QGA quantum enhancement, and the method comprises the steps: firstly, employing the true randomness and entanglement of quantum, achieving the coding of a corresponding relation between a task and a platform through a quantum state, and determining an optimal individual in a population through a fitness function; secondly, a quantum logic gate is used for achieving selection operation in the algorithm, and the rotation angle and direction of the quantum logic gate are determined through the fitness difference value between each chromosome and the optimal chromosome; then, a quantum logic gate X and a quantum logic gate SWAP are used for achieving mutation operation and crossover operation in the algorithm respectively; the method provided by the invention has high accuracy and timeliness, the number of algorithm iterations and the time complexity are greatly reduced, and the method can be effectively used for solving the task-platform optimization configuration problem in the military C2 organization.

The invention relates to the technical field of 3D action generation, and particularly provides a 3D human body action generation method and device, a storage medium, electronic equipment and a product, and the method can comprise the steps: obtaining a text feature vector corresponding to the text description of a human body action; layering and quantifying the text feature vector to obtain a basic action mark sequence and a residual action mark; acquiring a basic action sequence corresponding to the basic action mark sequence, and acquiring a residual action sequence corresponding to the residual action mark; and generating a 3D human body action sequence based on the basic action sequence and the residual action sequence. Some embodiments of the present application can improve the accuracy of 3D human motion generation.

The invention discloses a single-photon laser ranging method based on laser pulse frequency modulation deblurring, and belongs to the field of laser ranging. In the ranging process, two different pulse frequency references are firstly set to correspond to two coding code elements respectively, n-bit random codes are generated according to n-time non-fuzzy distance expansion requirements, it is guaranteed that the n-bit random codes do not have displacement symmetry, then a laser source is made to repeatedly emit corresponding pulse signals according to a sequence represented by the random codes, then echo signal sequences are received, and the pulse frequency references correspond to the two coding code elements respectively. And performing cross-correlation operation on the single photon accumulation histogram and a pulse sequence represented by a known random code to obtain an operation result, determining one of the single photon accumulation histograms with the maximum amplitude, and calculating a histogram time position corresponding to the result, namely, calculating a distance measurement result of de-blurring. According to the method, the problem of range ambiguity of pulse radar ranging can be solved, the time sliding sequence matching method proposed in cross-correlation operation has the advantages of being efficient and flexible, and the time complexity of cross-correlation operation is effectively reduced.

The invention discloses an underdetermined blind source separation method based on improved single source point detection, and the method comprises the steps: firstly, carrying out the initial single source point extraction of a mixed signal through employing the cosine angle criterion of a real part and an imaginary part of a time-frequency matrix column vector; then, abnormal points are removed according to whether the linear relation between the single source points and the abnormal points exists or not, the influence of the abnormal points on single source point detection is reduced, and therefore the estimation precision of a hybrid matrix is improved; then, clustering is carried out on the obtained single source points by using a K-means algorithm, so that an estimated hybrid matrix is obtained; and finally, recovering a source signal by using a least square method. The method provided by the invention not only considers the absolute direction of the real part and the imaginary part in the single source point, but also considers the linear relation between different single source points to remove the abnormal points, so that the estimation precision of the hybrid matrix can be improved, the noise immunity of the method is enhanced, and finally, a high-quality recovery signal is obtained.

The embodiment of the invention discloses a medicinal material replenishment method and device, electronic equipment and a computer readable medium. A specific embodiment of the method comprises the following steps: for each medicinal material in a target clinic, executing the following processing steps: acquiring a medicinal material ex-warehouse information sequence of the medicinal material in a preset time period; inputting the medicinal material ex-warehouse information sequence into a medicinal material replenishment information prediction model to obtain medicinal material replenishment predicted quantity information corresponding to a preset future time period; obtaining current stock value information of the medicinal materials and a preset warehousing replenishment total value; according to the inventory value information, generating current medicinal material unit value information of the medicinal material; determining actual medicinal material replenishment information corresponding to the medicinal materials according to the medicinal material unit value information, the medicinal material replenishment predicted quantity information and the warehousing replenishment total value; and according to the actual medicinal material replenishment information, the associated medicinal material transport vehicle is controlled to execute replenishment operation. According to the embodiment, the replenishment quantity of the medicinal materials can be accurately determined, repeated vehicle dispatching for medicinal material replenishment is not needed, and waste of transportation resources is avoided.

The invention discloses a metasurface multi-beam rapid generation method based on semi-supervision, and the operation process of the method comprises the following steps: 1, introducing an attention mechanism based on a convolutional neural network, and designing a cross entropy loss function; step 2, training the model by using a large number of samples obtained by caliber field superposition, migrating trained model parameters to a new model, and training based on the samples optimized by a traditional algorithm; step 3, semi-supervised learning is introduced, a model obtained through initial training is utilized to label unlabeled data samples, labeled sample data is added to label data samples, then the network model is retrained, the process is continuously repeated, and the number of the samples is iteratively increased; and 4, inputting required directional diagram data to obtain a corresponding coding matrix. Based on the method, a multi-beam coding problem is converted into a multi-label classification problem, the condition of repeated traversal in a codebook set in a traditional algorithm is avoided, and the time complexity is reduced.

The invention discloses a section output method, device and equipment of a three-dimensional model and a storage medium, and relates to the field of computer graphics, the section output method of the three-dimensional model comprises the steps that the three-dimensional model based on unit expression is acquired, and units are formed by connecting point data in the three-dimensional model; determining a binary space tree corresponding to the three-dimensional model based on a set maximum level, wherein each space node of the binary space tree comprises at least one unit; calculating a bounding box of each space node of the binary space tree, and screening out section units in the units according to positive and negative states of the bounding boxes and sections; and performing section calculation based on the section unit, and outputting a section model corresponding to the three-dimensional model. According to the method and the device, the technical effects of realizing section efficient performance calculation in a large model data set, eliminating a traversal calculation process of unnecessary units and greatly reducing calculation overhead can be realized.

The present invention provides a method for improving the positioning accuracy of an unmanned aerial vehicle for an observation target, monitoring a receiving end, obtaining first data, verifying the obtained first data, and then obtaining a real-time data set, filtering the data in the real-time data set, and obtaining first result data. By filtering the obtained first result data, a measurement point with the smallest error can be selected, and data with higher positioning accuracy can be obtained, thereby improving the accuracy of the positioning data, avoiding the inefficient process of analyzing a large amount of data one by one, and improving the measurement efficiency and positioning accuracy.

The embodiment of the invention discloses a movement control method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining grid maps of a plurality of levels, determining an initial starting point grid and an initial end point grid, then determining a target grid in a grid map of a target level, and then determining a target grid in the grid map of the initial level, the moving sub-path is determined in the mapping range of the target grid, and the space searched for determining the moving sub-path is small, so that the time complexity can be effectively reduced, the time consumption for determining the moving sub-path is reduced, and the smoothness of the target object navigation process is improved; the at least two moving sub-paths are determined in different time fragments, namely, the determination actions of the at least two moving sub-paths are split, so that the blockage phenomenon of the main thread is avoided, the smoothness of the navigation process of the target object is further improved, and the navigation efficiency of the target object is improved. The method can be widely applied to scenes such as cloud technology, artificial intelligence, games and intelligent transportation.

The invention discloses a WRSN charging method and system based on the lowest energy consumption of an unmanned aerial vehicle, and the method comprises the following steps: S1, taking the minimization of the total energy consumption of the unmanned aerial vehicle as a target, constructing a flight-hovering-based charging problem, and decomposing the flight-hovering-based charging problem into a minimization flight energy consumption problem and a minimization charging energy consumption problem; s2, aiming at the problem of minimizing charging energy consumption, providing an adaptive charging strategy based on flight and hovering charging modes, and carrying out joint optimization on the charging mode, the flight speed and the charging coverage radius when the unmanned aerial vehicle executes a charging task; s3, for a minimum flight energy consumption problem, decomposing the problem into an optimal non-charging flight speed determination problem and a minimum flight distance problem through a function decomposition mode; the optimal non-charging flight speed is obtained through a tangent method, and the problem of minimum flight distance is solved through a genetic algorithm. According to the invention, the total energy consumption of the unmanned aerial vehicle during the charging task can be effectively reduced, and the time complexity is relatively low.

The invention relates to the technical field of wiring, and discloses a wiring method, electronic equipment and a computer readable storage medium, and the method comprises the steps: converting an initial layout layout of a circuit board into a grid chart; according to the grid chart, a coarse wiring result of the initial layout layout is determined, and the coarse wiring result is an initial wiring path meeting a wiring task and achieving a wiring cross minimization target; generating a wiring geometric figure according to the safety spacing and the initial wiring path, and generating a device geometric figure according to the safety spacing and component information in the initial layout layout; and under the limitation of the geometric figure of the device, optimizing the position of the geometric figure of the wiring by taking the wiring constraint as a target to obtain a fine wiring result. According to the method, the wiring process is divided into a coarse wiring stage meeting the goals of connectivity and minimum wiring intersection number and a fine wiring stage meeting the wiring constraint, so that the solving complexity can be reduced, and a wiring scheme with the minimum wiring intersection number can be obtained in a relatively short time.

The invention provides a malicious encrypted traffic detection method based on behavior intention analysis, and belongs to the field of malicious traffic detection in encrypted traffic analysis. Comprising the following steps: providing a differentiated and robust traffic representation method called a traffic behavior matrix, dividing the whole traffic into time slices with fixed lengths, then calculating the accumulated packet lengths of outbound and inbound packets in each time slice, and combining the accumulated packet lengths into a matrix; according to the method, the classification model based on the convolutional neural network is adopted, the classification model learns the robust features of the traffic behavior matrix to ensure the effectiveness of malicious encrypted traffic detection, and malicious encrypted traffic detection in a real scene can be realized through the convolutional neural network model. The problems that malicious encrypted traffic is hidden in massive benign encrypted traffic in an existing real network scene and is difficult to detect, and the accuracy is low when an unbalanced data set is processed are solved.

The invention provides a path planning method of an inspection robot and a storage medium, and belongs to the technical field of path planning, and the method comprises the steps: taking the inspection robot only turning at a turning point in the inspection process as a target, and converting an inspection undirected graph into an inspection directed graph; determining a mapping point corresponding to each inspection point on the inspection directed graph, and determining an inspection point sequence of all the inspection points of the inspection robot by analyzing a topological relation between the mapping points; and in response to the issued current inspection task, determining the inspection sequence of the to-be-inspected points contained in the current inspection task according to the inspection point sequence, and realizing path planning of the current inspection task by using the inspection directed graph. According to the route planning method of the inspection robot, route planning of multiple target points of the inspection robot can be rapidly achieved according to the inspection requirement of a user, and the safety of the inspection robot and equipment in an inspection area is ensured by introducing a turning point turning method and the like.

The invention relates to a multi-sub-mirror structured short video batch mixing and cutting method with controllable repetition degree of materials, which comprises the following steps of: separately managing sub-mirrors and the materials, establishing a mapping relation between the materials and the videos, introducing a control value of an allowable maximum repetition sub-mirror number during calculation and combination, and performing gradient repetition degree operation from low to high. Through tracking and statistics of historical materials, a material minimum usage priority principle is applied, and video library repetition comparison is converted into material repetition comparison; the method has the advantages that the historical use records of the materials are persistent, and the use condition of each material can be tracked; the output video combinations are output according to the gradient of the repetition degree from low to high, the material repetition index of the material combination of each video is visualized, and the risk of preferential release is reduced to the minimum; the situation that all combinations are compared with all videos in a video library one by one is effectively avoided, and the time and space complexity of a combination calculation algorithm is reduced.

The invention discloses a finite-state machine coverage rate statistical method of a VCD (Video Compact Disc) file, which is characterized in that a mapping relationship is established between a signal name and state transition information by directly analyzing a YAML (Young Attribute Markup Language) format file of finite states and transition behaviors among the states compiled in a design stage and adopting a hash table mapping technology, so that the constant-level efficiency of reducing time complexity to O (1) is realized. When the VCD file generated by simulation is processed, two strategic file reading methods are adopted, and the file processing flow is simplified. And the uniqueness and accuracy of data processing are ensured through a self-defined Hash algorithm. In simulation data processing, screening is carried out according to the rule of the first character of the data, and only key state transition information is reserved. According to the method, the defects of processing the VCD file and counting the coverage rate of the finite-state machine under the condition of limited resources in the prior art can be effectively overcome, the counting speed of the coverage rate of the finite-state machine is increased, the efficiency of the verification process is improved, and remarkable performance improvement is brought to the verification stage of integrated circuit design.

The invention provides an efficient and stable wireless charging method suitable for a pressure-bearing lithium battery, and solves the problems that an existing maximum efficiency tracking control method has the defects of being high in time complexity, low in solving speed and low in maximum tracking speed, and the charging process is inaccurate due to the fact that parameters of the pressure-bearing lithium battery are affected by pressure. According to the invention, the improved perturbation observation-golden section algorithm is adopted to realize rapid maximum efficiency tracking and stable control of the pressure-bearing lithium battery. The method comprises the following steps: firstly, comparing a current solution with a previous solution and a next solution, and if the current solution is superior to the previous solution and the next solution at the same time, determining an interval in which an optimal solution is located; secondly, maximum efficiency tracking is carried out through a golden section search algorithm; the time complexity of the golden section search algorithm is generally O (log n), and n is the dimension of a solution space. Therefore, the high-efficiency stability of wireless charging of the pressure-bearing lithium battery can be improved to a certain extent by adopting the perturbation observation-golden section method.

A plant community multi-objective network optimization method in an intelligent logistics edge device completes intelligent logistics network multi-objective optimization by simulating behaviors of sowing, growing, flowering and fruiting of a plant community on the intelligent logistics edge device, and comprises the following steps: step 1, solving intelligent logistics network planning optimization by the plant community; step 2, optimizing a plant community solution intelligent logistics plan; step 3, performing optimization on plant community solution intelligent logistics execution; and step 4, optimization of the intelligent logistics reverse network for solving the plant community. According to the method, multi-objective optimization of the intelligent logistics network can be efficiently completed on the edge device.

The present invention discloses a method for detecting double loops of factor graphs, which obtains measurement information collected by multi-source sensors such as laser radar, satellite, and IMU, obtains state information based on an adaptive factor graph multi-source information fusion navigation algorithm, performs coarse loop detection based on state constraints, and if it passes, uses the laser radar point cloud map to perform precise loop detection. If both double loop detections pass, the state information is extracted to perform factor graph loop correction; if it fails, the next time node is detected. The present invention can be applied to scenes such as urban positioning, vehicle speed measurement, and attitude alignment; the present invention has good robustness and low time complexity, and can shorten the factor graph loop detection time.

The invention discloses a training method and device for a federal learning decision tree model, and relates to the technical field of big data. A specific embodiment of the method comprises the following steps: acquiring first credit characteristic data and a credit label of each user in a plurality of users; determining a first dissimilarity matrix according to the first credit feature data; receiving a second dissimilarity matrix sent by the data provider; determining global perception sample selection probability distribution according to the first dissimilarity matrix and the second dissimilarity matrix; and then cyclically executing the following processes: training to obtain a decision tree model: determining a user core set of the current node according to global perception sample selection probability distribution, then determining each sub-node after the current node is split according to the user core set, and taking each sub-node as the current node. According to the embodiment, the accuracy of model prediction can be improved, the performance of the model is improved, the optimization efficiency of the model and the training speed of the model are improved, and the training cost of the model is reduced.

The invention provides an epilepsy prediction method and system based on a 2D-CNN and a Cox-Stuart early stop mechanism, and the method comprises the steps: obtaining electroencephalogram data; preprocessing the electroencephalogram data, wherein the preprocessing comprises normalization and filtering; inputting the preprocessed signal into a pre-trained convolutional neural network epilepsy prediction model for feature extraction and classification, and predicting to obtain an epileptic seizure state; the training process of the pre-trained convolutional neural network epilepsy prediction model comprises the steps of putting training set data and verification set data into a CNN model for training and verification; judging whether a loss function of the verification set has a rising trend or not by adopting a Cos-Stuart test method, and early stopping; and storing the trained model parameters, and testing the model performance by using the test set data to obtain the pre-trained convolutional neural network epilepsy prediction model. Under the condition that only the data is preprocessed, the time domain signal of the data is directly used as the input of the model, so that the time complexity and the calculation complexity are reduced.

The invention provides a rapid maximum efficiency tracking control method and system for a wireless charging system, and solves the problems of high time complexity, low solving speed and low maximum tracking speed of an existing maximum efficiency tracking control method. According to the method adopted by the invention, firstly, a current solution is compared with a previous solution and a latter solution, and if the current solution is superior to the previous solution and the latter solution at the same time, an interval in which an optimal solution is located can be determined; secondly, maximum efficiency tracking is carried out through a golden section search algorithm; the time complexity of the golden section search algorithm is generally O (log n), and n is the dimension of a solution space. The solution space dimension is generally greater than 1, and O (log n) is less than O (n) at the moment, so that the maximum efficiency tracking speed can be improved to a certain extent by adopting the perturbation observation-golden section method.

This invention belongs to the technical field of embedded artificial intelligence and real-time system scheduling, and more specifically, relates to an adaptive load resource scheduling method and apparatus for embedded GPUs. The method involves offline acquisition and modeling of the platform's thermal throttling behavior, constructing DNN variants with different computational loads and accuracies, and profiling them under multiple frequency and TPC configurations to obtain a latency-power mapping table. In the online phase, based on task deadline margins and current temperature / frequency states, the method looks up the variant with the highest accuracy and meeting the deadline under the current thermal conditions and selects the corresponding TPC configuration from the profiling table. If necessary, phased online pruning and low-overhead switching are implemented, and the method collaborates with system-level DVFS / task mapping to optimize energy efficiency. This method can still guarantee the real-time performance of inference tasks under passive frequency reduction conditions and significantly reduce performance fluctuations caused by thermal throttling.

According to the image Smear rapid elimination method and device for the satellite load ground detection platform, after a detection platform receives a data file of an image, a formula (12) is firstly used for calculating a Smear component of each column, operation of a formula (13) is executed, and then a true value of each pixel after Smear elimination can be obtained. Compared with a matrix elimination algorithm, the algorithm of the image Smear rapid elimination method for the satellite load ground detection platform has the advantages that the algorithm is high in operation speed by counting the operation time of a program, and the correction efficiency of the Smear effect can be greatly improved. The algorithm provided by the invention can be directly embedded into control software of a ground inspection station, and compared with a traditional offline Smear elimination algorithm, the algorithm provided by the invention can realize online real-time Smear elimination, so that technicians can conveniently correct optical and electrical parameters of an imaging spectrometer in time.