46results about How to "Shorten convergence time" patented technology

The invention discloses a charging and discharging control system and electronic equipment. The method comprises the following steps: S1, constructing a charging pile multi-target optimization model; s2, initializing a chaotic particle swarm optimization algorithm; s3, each particle in the particle swarm is optimized according to an objective function in the charging pile multi-objective optimization model; s4, carrying out loop iteration until the chaotic particle swarm optimization algorithm reaches a convergence condition; s5, generating a prediction result of the operating environment of the charging pile; s6, updating the current optimal particle and the global optimal particle; and S7, generating a multi-target charging and discharging optimization control strategy of the charging pile according to the optimized global optimal particle, applying the optimization control strategy to a charging pile charging and discharging control execution module, and executing multi-target charging and discharging control of the charging pile. According to the method, the globally optimal solution in a complex scene can be quickly solved, and the efficient operation requirement of the electric vehicle charging pile in a large-scale deployment scene is met.

The invention provides a text classification method based on quantum transfer learning, and belongs to the field of fusion of quantum computing and artificial intelligence. In consideration of complex feature space and nonlinear characteristics in text classification, quantum circuit design, a self-attention mechanism and a quantum transfer learning process are optimized, so that the learning efficiency and classification accuracy of the model are improved. The high-dimensional feature representation and a complex nonlinear decision process are converted into a quantum self-attention network structure, an optimal strategy of text classification is obtained by utilizing a complex quantum kernel self-attention network (CQKSAN) model, and a quantum superposition and entanglement mechanism is introduced to enhance the expression ability of the model. A simulation result shows that compared with a traditional deep learning algorithm and other mixed quantum-classical algorithms, the algorithm provided by the invention has higher classification precision and learning efficiency. The method not only shows the potential of a mixed quantum-classical method, but also provides a basis for further research of a future full-quantum learning system.

The invention discloses a PID (Proportion Integration Differentiation) parameter optimization method based on an improved graffiti search algorithm, and relates to the technical field of PID. The invention aims to solve the problems of insufficient local search capability, slow convergence speed and poor system stability in the prior art, thereby realizing a better PID parameter optimization effect. Specifically, the PID parameter optimization method based on the improved graffiti search algorithm comprises the following steps: constructing a PID controller model; a step of replacing a fixed perception probability and a step length in the model with an adaptive perception probability and a Levy flight step length; a step of adding a crow misleading prevention factor into the model; and acquiring preset parameters in the PID controller model based on a particle swarm algorithm. The preset parameters comprise proportional (Kp), integral (Ki) and differential (Kd) parameters. Parameters of the Levy flight step length are dynamically adjusted through a Bayesian optimization algorithm. The method is suitable for being applied to PID parameter optimization work.

The invention discloses a defect detection system, method and device based on artificial intelligence image processing, and particularly relates to the field of image defect detection.A data acquisition module is used for obtaining image data from industrial equipment, a sensor and a camera and conducting preliminary formatting processing on the image data. The model convergence time is shortened, the utilization efficiency of training resources is improved, the generalization ability of the model is enhanced, and the performance of the model in processing a rare scene is improved; according to the method, high accuracy can be kept while computing resource consumption can be remarkably reduced, it is ensured that the defect detection system can also operate efficiently in resource-limited equipment, different hardware conditions are adapted, the application scene requirement of the system is expanded, the operation problem caused by environmental conflicts is reduced, the system safety is enhanced, and the system performance is improved. And the influence on system operation caused by external malicious software attack or resource leakage is avoided.

The invention discloses a beam focusing optimization method based on dense connection expansion self-coding element learning, and relates to the technical field of super-large-scale IRS assistance, and the method comprises the following steps: S1, analyzing a super-large-scale intelligent reflector-assisted multi-user communication system, and obtaining a problem optimization formula; s2, constructing a beam focusing optimization algorithm network based on dense connection expansion self-coding element learning according to a problem optimization formula, and taking a matrix form of a phase shift matrix and a precoding matrix as the input of the beam focusing optimization algorithm network based on dense connection expansion self-coding element learning, and obtaining an optimized pre-coding matrix and an optimized phase shift matrix, and obtaining the maximum spectrum efficiency value of the super-large scale intelligent reflector assisted near field communication system according to the optimized pre-coding matrix and the optimized phase shift matrix. According to the invention, the spectrum efficiency of the system can be improved and the processing overhead of high-dimensional signals can be reduced.

The invention discloses an autonomous underwater robot system over-torque sliding mode control trajectory tracking method based on an improved adaptive law. The method comprises the following steps: establishing a six-degree-of-freedom autonomous underwater robot dynamic model; converting the six-degree-of-freedom autonomous underwater robot dynamical model, and constructing a six-degree-of-freedom autonomous underwater robot dynamical model based on lumped disturbance; designing a synchronization time estimator to estimate the lumped disturbance in the six-degree-of-freedom autonomous underwater robot kinetic model based on the lumped disturbance; and based on a sliding mode surface, a synchronization time estimator and a hyperbolic tangent function, designing a super-twist sliding mode controller based on an improved adaptive law, and realizing trajectory tracking of the autonomous underwater robot. According to the invention, buffeting of the system can be effectively suppressed while rapid convergence of the system is ensured.

The invention discloses a mechanical arm path planning method and system based on self-adaptive expansion sampling, belongs to the technical field of path planning, and solves the problems of low sampling rate, low convergence rate and poor initial path quality caused by too random sampling area and improper selection of expansion step length during path planning of a mechanical arm in the prior art. According to the self-adaptive expansion sampling strategy provided by the invention, on the basis of the F-RRT * optimization algorithm, the growth direction of the random tree is corrected by introducing the target bias force, so that the random tree deviates to the target point, the inflection point is optimized, the generated path is smoother, and by introducing the concept of target point gravity deviation, the optimal path is obtained. A strategy is adaptively expanded according to the obstacle complexity in the environment, and expansion is performed by dynamic step length, so that the stability and precision of the trajectory are improved, and the length of the generated path, the sampling frequency and the convergence time are reduced.

The invention relates to the field of electric digital data processing, in particular to a multi-mode integrated rainfall prediction method based on artificial intelligence. The method comprises the steps of collecting meteorological data, training a neural network based on chaotic disturbance, training an LSTM based on quantum Monte Carlo, collecting new meteorological data, and inputting new precipitation influence factor features into the trained neural network based on chaotic disturbance for feature fusion. And inputting the actual feature vector set into the trained quantum Monte Carlo-based LSTM for prediction to obtain the precipitation of the target date. The existing multi-mode integrated rainfall prediction method has the problem of low prediction accuracy. The multi-mode integrated rainfall prediction method based on artificial intelligence provided by the invention is relatively high in prediction accuracy.

The present invention discloses a fighter flight control method based on adversarial skill embedding and hierarchical reinforcement learning. The method designs an adversarial skill embedding learning framework composed of an adversarial generative algorithm and a deep reinforcement learning algorithm. The method uses expert flight data to automatically learn the mapping of underlying control instructions to macro maneuvering skills by an AI algorithm. By reusing these macro maneuvering skills, the method significantly reduces the learning difficulty of downstream air combat tasks, greatly compresses the convergence time of the reinforcement learning algorithm, improves the game performance of air combat intelligent agents, and gives air combat intelligent agents higher flexibility and rapid migration capabilities, thereby exploring a new technical path for the practical training and deployment of air combat intelligent agents.

The invention provides a positioning method and device based on PPP-B2b, and belongs to the field of satellite navigation and positioning. The positioning method comprises the steps that firstly, pseudo-range observation data and phase observation data of a BDS satellite and a GPS satellite and PPP-B2b precise correction information are obtained, and the PPP-B2b precise correction information comprises satellite precise orbit parameters, a satellite clock difference reference and initial satellite clock difference deviation between the satellite clock difference reference and each satellite; the method comprises the following steps: setting a receiver clock error parameter in a BDS / GPS observation model based on a satellite clock error reference, taking an initial satellite clock error deviation as a variable added for a pseudo-range observation equation of each satellite in the BDS / GPS observation model, and establishing a precision positioning mathematical model based on PPP-B2b service; and solving the precision positioning mathematical model according to the pseudo-range observation data and the phase observation data of the BDS and the GPS satellites to obtain a precision positioning result of the receiver. According to the method, the precision and reliability of BDS / GPS precision positioning based on PPP-B2b are improved, and the convergence time is saved.

The invention provides a distributed steer-by-wire vehicle fault-tolerant control method. The method comprises the steps that a distributed steer-by-wire vehicle dynamics model, a two-degree-of-freedom reference model and a steering motor failure model are established; in the design of the upper fault-tolerant controller, a self-adaptive terminal sliding mode control method is adopted, the method has the characteristics of high robustness and self-adaptive parameter change, and the robustness of the system can be considered while the target tracking performance is ensured. In the design of a lower-layer torque and rotation angle optimal distribution reconstruction controller, a failure factor is introduced, two more objective functions are established on the basis of minimizing the tire load rate, and errors between longitudinal and lateral forces and expected target values and errors between yawing moment and additional yawing moment are reduced while the optimized objective functions are ensured to have solutions. According to the method, the uncertainty of the system during the steering failure is fully considered, and the robustness of the system can be considered while the target tracking performance is ensured, so that the control stability of the distributed steer-by-wire vehicle during the steering failure is improved.

The invention relates to a wireless channel multipath signal parameter joint estimation method based on variational Bayesian inference, and the method comprises the steps: carrying out the modeling of an amplitude gain through employing Bernoulli Gaussian level prior under a framework of VBI space generalized alternating iteration solving, and obtaining a pruning condition expression; a time delay parameter pre-estimation expression is obtained based on the self-correlation characteristic of observation sending signals, in each iteration process, time delay parameters under the current mirror component number are estimated according to the time delay parameter pre-estimation expression, then component estimation and dispersion parameter estimation of one mirror component are calculated, and the dispersion parameter estimation of the mirror component is calculated. And judging whether a pruning condition expression is met or not, if so, calculating amplitude gain parameter estimation, if not, cutting off the mirror component, and alternately solving the parameter estimation of the next mirror component until convergence to obtain the number of the mirror components and the parameter estimation of each mirror component. According to the method, the accuracy of the result can be ensured while the solving efficiency is effectively improved.

This invention provides a fast Doppler localization method based on opportunistic signals. The method includes: establishing instantaneous Doppler frequency shift measurements for each satellite based on the receiver's initial position and velocity information, the wavelength of the opportunistic signal, and the position and velocity information of each satellite; establishing an instantaneous Doppler localization equation based on the relationship between the instantaneous Doppler frequency shift measurements for each satellite and the receiver's position information; determining regularization parameters using the variance expansion factor method; obtaining constraint parameters based on the regularization parameters and the stable functional; and solving the instantaneous Doppler localization equation under the condition of minimizing the constraint parameters to obtain the instantaneous Doppler localization result based on the opportunistic signal. The localization result includes the receiver's final position information, final velocity information, and frequency offset.

The invention discloses a multi-pose image data generation method and device for face recognition, equipment and a storage medium, and relates to the technical field of face image generation. The method comprises the following steps: obtaining a face data set of a specific scene and adjusting a face image generation model to obtain a new generation model; generating a new face data set by using the new model, and extracting feature vectors through a potential vector generator to form a potential vector set; performing face detection and screening on the new data set, classifying potential vectors according to a posture estimation angle, and labeling an angle value as a label; training the potential vector with the label by using a support vector machine to obtain decision boundaries of different attitude angles; and performing linear interpolation on the potential vector based on the decision boundary, and generating face images at different attitude angles by using the generation model. According to the method, more accurate and robust face image generation is realized, and the method has important significance in improving the accuracy and efficiency of multi-pose face recognition.

The embodiment of the invention discloses a routing method and device for data transmission, equipment, a medium and a program product, which are used for simplifying the data routing complexity, reducing the calculation overhead and improving the data routing performance in a high dynamic environment. The method comprises the steps that a data transmission notification is received, the data transmission notification comprises a source node and a destination node for data transmission, one of the source node and the destination node is a ground network element, and the other one is a satellite-borne network element; according to pre-maintained satellite-ground topological information and inter-satellite topological information, routing information of each routing segment in a plurality of routing segments between the ground network element and the satellite-borne network element which are configured in advance is calculated, and the satellite-ground topological information comprises a connection relation between a gateway station and a satellite node; the inter-satellite topological information comprises a connection relationship among a plurality of satellite nodes; and performing routing configuration according to the routing information of each routing segment.

The application relates to the technical field of chip design, in particular to an AI-based DDR5 memory interface chip adaptive method and system.The method comprises the following steps: obtaining a configuration instruction stream, extracting current and historical configuration instructions, and predicting target pins, switching time and target modes.Before the switching time, target electrical parameters are calculated according to pin electrical characteristics, historical switching records and channel load characteristics.During the switching, the target parameters are loaded and the temporary impedance matching network is switched in.After the switching, a fine-tuning increment is generated according to the deviation of the signal quality and the threshold value, and the updated electrical parameters are obtained by superposition, and the matching network is disconnected after the updated electrical parameters are written back to the register.According to the timing error and the fine-tuning increment generated by multiple switching, the weight parameters are updated and deployed back to the chip.Through AI prediction and the dynamic switching and fine-tuning of the temporary impedance matching network, seamless, stable and low-delay mode switching of the memory interface chip in the reconfigurable computing system is realized.

The application discloses a projection figure distortion prediction compensation correction method of a laser scanning projection system, and comprises the following steps: carrying out laser scanning projection system projection; acquiring projection figure distortion data, establishing a neural network training data set and a test data set; establishing an initial topological structure, weights and thresholds of a BP neural network; optimizing the weights and thresholds of the BP neural network based on a particle swarm optimization algorithm; carrying out deviation compensation on the coordinate positions of each projection point according to the training output of the PSO-BP neural network; selecting feature points in each part of a to-be-projected plane; projecting the feature points by using a distortion deviation correction projector based on the PSO-BP neural network; acquiring feature point projection coordinate values; subtracting the ideal projection point coordinates without distortion from the corrected feature point projection coordinates; judging whether the difference value meets an error interval; if yes, it is determined that the projection coordinate value distortion error has been corrected and compensated; if no, returning to step three for feedback type multiple iterations until the difference value meets the error interval.

The application belongs to the technical field of wireless communication, and specifically provides a wideband analog adaptive cancellation circuit based on digital accumulation loop iteration, which is used to solve the problem of reduced iteration depth and decreased cancellation ratio of the existing analog adaptive cancellation circuit due to the incomplete digital integral function. The application comprises an analog adaptive cancellation module and a digital enhancement module, taking the analog adaptive interference cancellation module as the main body, using the digital enhancement module (digital circuit) to process the direct current component, and realizing the accumulation function when updating the weight, i.e. completely realizing the digital integral function, which significantly increases the iteration depth of the system and improves the cancellation ratio of the system. Meanwhile, the algorithm of the digital enhancement module can be adjusted to determine the convergence condition of the system, and the convergence time is reduced under the condition of meeting the demand cancellation ratio. Furthermore, the analog phase shifter is used to change the optimal cancellation frequency of the system, increase the working bandwidth of the system, and relax the convergence condition of the system.

The invention relates to the technical field of permanent magnet synchronous motor control, and discloses a permanent magnet synchronous motor sliding mode control method, device and equipment and a storage medium, and the method comprises the steps: constructing a motor state equation based on a permanent magnet synchronous motor mathematical model, taking a load torque as external disturbance, and constructing a novel adaptive nonlinear expansion state observer; based on a motor state equation, constructing a non-singular fast terminal sliding mode surface and a novel variable gain index reaching law; constructing a novel non-singular fast terminal sliding mode controller based on the novel adaptive nonlinear extended state observer, the non-singular fast terminal sliding mode surface and the novel variable gain index reaching law; and executing permanent magnet synchronous motor sliding mode control by using the novel nonsingular fast terminal sliding mode controller. Compared with the prior art, the novel sliding mode control strategy has the advantages that buffeting is greatly weakened, the control precision is improved, the convergence speed is obviously increased, and the system can be asymptotically converged to a balance point within finite time.

The invention discloses a non-intrusive load monitoring method and device for house energy consumption and a storage medium, and belongs to the technical field of power monitoring, and the method comprises the following steps: carrying out the random initialization of all particles through employing a dynamic particle swarm optimization algorithm, and obtaining a particle hyper-parameter combination; generating a long short-term memory network model; dividing the obtained historical power classification data set into a training set and a test set according to timestamps, training through the training set, and calculating through the test set to obtain a root-mean-square difference; using the root-mean-square difference as an adaptive value to update the individual optimal solution of the particles and the current global optimal solution of the whole particle swarm, then calculating the weight of the particles, and updating the speed and the position according to the weight of the particles; determining the optimal solution of the iteration complete bureau as the optimal parameter configuration of the long short-term memory network model; and inputting the real-time power data to obtain a load monitoring result. The method can solve the problems that an existing non-intrusive load monitoring method is high in complexity, long in consumed time and low in monitoring result accuracy.

The invention discloses a photovoltaic MPPT (Maximum Power Point Tracking) control algorithm, which relates to the technical field of photovoltaic control algorithms, utilizes the characteristics of simple structure, high convergence speed, high global optimization capability, wide applicability and the like of a gradient optimization algorithm, and improves the defects of premature convergence and sensitivity to control parameters of the gradient optimization algorithm. Meanwhile, the photovoltaic output energy efficiency and the stability of the photovoltaic output energy efficiency serve as the evaluation basis, an MPPT algorithm performance evaluation model is constructed through an analytic hierarchy process, photovoltaic MPPT performance is evaluated, and GTO performance is verified and improved.

The invention discloses an ad hoc network dynamic topology routing optimization method and system based on a Deepseek lightweight large model, and the method specifically comprises the steps: collecting the information of each node in an ad hoc network, and standardizing the node information into a feature vector format which can be analyzed by the Deepseek model; the standardized node feature vectors are input into a Deepseek model, and the topological structure change trend in a short time in the future is predicted; generating a routing path selection strategy based on a topology change trend predicted by the model, testing in an ad hoc network simulation platform, and adjusting feature weight distribution of the model according to a test result; and deploying the adjusted strategy to an actual network, collecting routing performance data in real time as a new sample input model, iteratively updating the model, and improving the accuracy of topology prediction and routing optimization. According to the method, a response can be quickly made to a new network topology, the convergence speed is increased, accurate routing selection can be realized, node energy consumption is reduced, and the life cycle of the ad hoc network is prolonged.

The present invention proposes a charging information transmission method for mobile new energy vehicles in the Internet of Vehicles environment. First, by using distributed mobile edge computing and available vehicle-mounted ad hoc networks, an efficient heterogeneous communication framework based on hybrid transmission mode is designed for large-scale charging information transmission between mobile new energy vehicles and charging stations; secondly, based on the constructed terminal intersection model and ant colony algorithm, an adaptive routing protocol suitable for the Internet of Vehicles environment is proposed to dynamically plan the optimal charging information transmission path; finally, by using the constructed waiting time model, a charging information forwarding mechanism based on the receiving end is proposed to select the best charging information forwarding vehicle. The present invention has been experimentally verified that the proposed charging information transmission method can effectively reduce wireless network resource overhead, alleviate network congestion, and improve the adaptability of charging data transmission.

The present invention relates to the field of electronic digital data processing, and in particular to a multi-mode integrated precipitation prediction method based on artificial intelligence. The method comprises collecting meteorological data, training a neural network based on chaos perturbation, training an LSTM based on quantum Monte Carlo, collecting new meteorological data, inputting new precipitation influencing factor features into the trained neural network based on chaos perturbation for feature fusion, inputting an actual feature vector set into the trained LSTM based on quantum Monte Carlo for prediction, and obtaining the precipitation on the target date. The existing multi-mode integrated precipitation prediction method has the problem of low prediction accuracy. The multi-mode integrated precipitation prediction method based on artificial intelligence provided by the present invention has high prediction accuracy.

This invention discloses a routing partitioning method and apparatus for large-scale low-Earth orbit satellite networks, primarily addressing the problems of high link overhead and slow convergence speed in existing satellite networks. The scheme includes: generating a spatial network topology based on constellation size, number of orbital planes, orbital altitude, and orbital inclination; dividing the constellation into multiple routing regions based on the relative positions of the satellites, establishing link state information exchange rules, a routing table storage and update mechanism, and a data packet forwarding mechanism; establishing a quantitative relationship between the number of partitions and link overhead based on the network topology, routing region partitioning results, and link state information exchange rules; establishing a quantitative relationship between the number of partitions and routing convergence time based on the dynamic topology change characteristics of the satellites, routing region partitioning results, and routing table storage and update mechanism; calculating link overhead and routing convergence time under different numbers of partitions, constructing routing performance indicators, and outputting the optimal routing partitioning result. This invention can reduce link overhead and accelerate routing convergence speed, and can be applied to large-scale satellite network communication systems.

According to the RGB-D saliency target detection method based on mixed feature cross fusion, through a mixed feature cross fusion module, a model can fully utilize complementary information between an RGB image and a depth image, and efficient fusion of features is achieved. The module enhances the interaction between RGB and depth features through a self-attention mechanism, so that the connection between different modal features is better modeled and captured, and the problem of feature distortion under the condition of poor depth image quality is solved. Besides, the invention provides a specific training strategy and an optimization scheme, and the model can rapidly converge on a large-scale data set by adopting the combination of an MAE pre-training model and a ViT model, and obtains excellent performance in a saliency target detection task.

The present invention discloses a chlorophyll content estimation method based on optical characteristic parameter inversion, which comprises the following steps: (1) constructing a leaf model based on a surface light source to simulate photon transmission and obtain a simulated spectral image; (2) inputting the simulated spectral image into a convolutional neural network for training and obtaining a pre-trained model; (3) based on the pre-trained model, performing maximum mean difference (MMD) migration on measured spectral data and obtaining an optical parameter inversion model of all training layers; (4) using the optical parameter inversion model obtained by MMD migration to estimate the chlorophyll content of the leaf. The optical parameter inversion method of green radish leaf based on MMD migration has better performance than the common model migration method.

A mechanical arm path planning method and system based on adaptive expansion sampling belong to the technical field of path planning, and solve the problems of slow sampling rate, slow convergence speed and poor initial path quality caused by too random sampling area and unsuitable expansion step length selection when planning the path of the mechanical arm under the prior art; the adaptive expansion sampling strategy provided by the present application introduces a target bias force to correct the growth direction of the random tree based on the F-RRT* optimization algorithm, so that it is biased towards the target point, optimizes the inflection point, makes the generated path smoother, introduces the concept of target point gravity deviation, adaptively expands the strategy according to the complexity of obstacles in the environment, and expands with a dynamic step length, improves the stability and precision of the trajectory, and reduces the length of the generated path, the sampling times and the convergence time.

The invention discloses an autonomous cooperation method and system for a low-altitude emergency federated learning client, and aims to solve the problem of training efficiency and resource optimization of an unmanned aerial vehicle group in a complex environment. The method comprises the following steps: firstly, carrying out offline reinforcement learning on a pre-training strategy network through a simulation environment, and implementing a forced participation mechanism at the initial stage of deployment to accumulate real experience; secondly, each unmanned aerial vehicle encodes a local state into a message vector, exchanges and fuses with a neighbor to form a cooperative intention vector, and autonomously determines an action probability of participating in federated learning; then, in a training stage, the selected client executes local training and uploads update through differential gradient compression, and the server performs weighted aggregation according to gradient difference and a precision improvement value. And finally, performing strategy updating of centralized training and distributed execution by using the QMIX hybrid network according to the multi-target reward function, and realizing closed-loop collaborative optimization.

The invention discloses a GNSS anti-spoofing detection method and system based on Doppler multiple check, and belongs to the technical field of satellite navigation safety, and the method comprises the steps: obtaining and preprocessing multi-frequency GNSS observation data through adaptive protocol analysis; the historical state is updated by adopting a main frequency point anchoring strategy, and the group delay influence is isolated; single-frequency pseudo-range rate and Doppler consistency verification and multi-frequency-point Doppler Z-Score coherence cross validation are executed in parallel, and deception signals are identified from two dimensions of a physical layer and an inter-frequency relation; a historical state self-cleaning mechanism driven by a state machine is introduced, rapid resetting is achieved when the signal environment changes drastically, and misinformation is avoided; and finally outputting a global alarm according to an abnormal satellite statistical result. The method can effectively resist advanced generative deception jamming, realizes millisecond-level rapid recovery after jamming is finished, has good hardware compatibility and integration, and is suitable for scenes such as unmanned aerial vehicles, vehicle-mounted terminals and the like which have strict requirements on real-time performance and power consumption.