142 results about "Learning dynamics" patented technology

The invention belongs to the technical field of power equipment, and discloses a power equipment fault early warning method based on multi-source data fusion, and the method comprises the steps: constructing multi-dimensional feature association through multi-modal data time-space association collection and hierarchical fusion driven by a knowledge graph; a space-time weight matrix is used for correcting sampling deviation, fault mechanism knowledge is combined to strengthen key feature contribution degree, false alarm and missing alarm caused by data isolation are effectively avoided, early recognition of hidden defects of equipment is realized, and global perception capability of early warning is improved. A meta-learning enhanced cross-equipment early warning model and reinforcement learning dynamic threshold decision are adopted, cross-equipment rapid adaptation under a small number of samples is realized through a ''meta-micro'' double-circulation mechanism, and a nonlinear law of fault evolution can be accurately described by combining a three-dimensional dynamic threshold matrix to balance an equipment state, an environment and an operation and maintenance strategy. The model generalization problem of different types of equipment in a complex environment is solved, and the adaptability to scenes such as load fluctuation and environment sudden change is improved.

The invention discloses an intelligent optimization method for a multi-type well seam joint control fine injection-production mode, and relates to the technical field of oil-gas field development. The method comprises the following steps: setting a well seam joint control fine injection-production mode, establishing an oil reservoir numerical simulation model in oil reservoir numerical simulation software, obtaining multiple groups of oil reservoir injection-production schemes based on a Latin hypercube sampling method, performing simulation according to each group of oil reservoir injection-production schemes by utilizing the oil reservoir numerical simulation model, generating multiple pieces of sample data, and establishing a sample database; a deep learning agent model is established, after the sample database is utilized to train and train the deep learning agent model, a particle swarm optimization algorithm is adopted to carry out single-target pre-search global optimization to obtain a preferred reference strategy, a reinforcement learning dynamic decision model is established, and a reinforcement learning agent is obtained through training based on a PPO near-end strategy optimization algorithm; and the optimal injection-production development scheme of the oil reservoir is obtained by utilizing the reinforcement learning agent, so that rapid optimization and decision support of the oil reservoir injection-production scheme in a new multi-type well seam joint control mode are realized.

The invention relates to an intelligent drainage basin management method, device and equipment based on digital twinning and a medium of the intelligent drainage basin management method and device based on digital twinning. The method comprises the steps that core terrain attributes are extracted through terrain feature decoupling and coding, and a disaster response function of an optimal source drainage basin is migrated based on terrain and rainfall similarity; a hydrodynamic model and a migration function are combined to generate a physically constrained synthetic disaster situation data set, a two-channel neural network model fusing topographic features and rainfall dynamics is constructed, and model parameters are dynamically calibrated through Bayesian continuous learning by using real-time monitoring data of a target drainage basin. Finally, a digital twinborn system with flood routing prediction capability is formed, the problem of rapid construction of a flood prediction model under the condition of no historical data is solved, and the timeliness and accuracy of early warning of sudden flood in small and medium-sized watersheds are remarkably improved.

The invention discloses a hybrid sequential network power consumption prediction method for reinforcement learning dynamic calibration, and the method comprises the steps: collecting multivariable power consumption time sequence data, completing the data preprocessing through resampling, feature engineering, normalization and sliding window technologies, generating a supervised learning sample set, and dividing the supervised learning sample set into a training set, a verification set and a test set; constructing a hybrid prediction model comprising a dynamic capture module, a long-term dependence modeling module, a regression prediction module and a reinforcement learning dynamic fine tuning module; training and optimizing by adopting a staged training strategy to obtain a hybrid prediction model; multivariable power consumption time sequence data are collected in real time and preprocessed, the preprocessed data serve as input, real-time prediction of future total consumption is conducted through the mixed prediction model, and a final prediction result after dynamic fine adjustment is output. According to the method, accurate and efficient prediction of the power grid load can be realized, and a reliable technical solution can be provided for power system scheduling optimization, demand side management, market transaction and other scenes.

The invention discloses a global scanning image splicing and defect positioning system for an ultra-large-diameter quartz crucible, and relates to the technical field of quartz crucibles, and the system comprises a scanning control module which constructs a fusion model through multiple sensors, evaluates a risk coefficient, obtains the fusion model, dynamically plans a multi-arm path of a robot based on reinforcement learning, and sends the multi-arm path of the robot; adaptively adjusting the acquisition frequency of the corresponding area based on the risk coefficient, and outputting multi-modal data through a multi-round feedback iteration optimization acquisition process; the image splicing module is used for acquiring multi-modal data for local splicing, and generating a panoramic image in combination with a multi-arm path of the robot; the defect positioning module is used for constructing a quartz crucible model through the panoramic image, detecting defects in a layered manner by utilizing a YOLOv5 model, tracking the defects through extended state Kalman filtering, quantifying the defect volume and the hazard level by combining point cloud semantic segmentation, and performing 3D positioning and evaluation on the defects; the output data is ensured to always meet the subsequent image splicing and defect positioning requirements, and the system stability is guaranteed.

The invention provides a power grid standard intelligent recommendation method based on reinforcement learning and post system management. The power grid standard intelligent recommendation method comprises the following steps of 1, collecting and cleaning power grid standard data; step 2, post-standard system knowledge graph construction; 3, designing a reward function for the power grid standard recommendation system, and realizing a personalized recommendation strategy by taking post responsibilities and skill levels as dynamic weight parameters; and step 4, post-driven online incremental training is carried out. According to the method, a reinforcement learning dynamic optimization engine is adopted to learn user behaviors (click rate and task completion score) in real time, and a standard recommendation strategy is adjusted; and when the standard is updated, the post capability weight is automatically adjusted, closed-loop feedback optimization is realized, and the problem of updating delay of a traditional system is solved.

The invention discloses a power grid energy storage demand response multi-agent reinforcement learning dynamic optimization method, which belongs to the technical field of power system scheduling and optimization control, and comprises the following steps of: constructing a joint optimization model based on a Markov decision process, representing an energy storage charge state, demand response regulation potential and renewable energy output fluctuation, defining a state space, and establishing a dynamic optimization model; establishing a joint action space; constructing a dual-network structure; an improved hierarchical time memory network is introduced, spatial-temporal features are extracted through a multi-level memory unit and a spatial-temporal attention mechanism, and the spatial-temporal features are used for strategy optimization and value evaluation; optimization learning is carried out by adopting multi-agent interaction and centralized training decentralized execution, and strategy updating is carried out by combining an experience playback pool and target network soft updating; and outputting a power grid dynamic scheduling strategy, realizing energy storage and demand response joint optimization, and improving system stability and adaptive capacity under a high-proportion renewable energy access condition.

The invention provides a computer network data information identification system, which relates to the technical field of computer network security and data processing and comprises an edge cloud collaborative data acquisition module, a multi-modal feature fusion module, a federated learning dynamic model training module and an intelligent decision and response module. The method has the advantages that the edge cloud collaborative architecture is adopted, data preprocessing and feature extraction are conducted on the network edge, the transmission quantity and delay are reduced, and the real-time performance and the processing efficiency are improved; a self-attention mechanism is used for fusing multi-modal features, so that the recognition accuracy is improved; the data privacy security is protected based on a federated learning framework training model; a reinforcement learning algorithm is introduced to optimize federated learning and decision strategies, and the adaptability and intelligence of the system are enhanced; and response abnormal data can be intelligently dispatched according to an identification result to ensure safe and stable operation of the network.

The invention discloses an efficient acquisition and processing system for safety production monitoring data of an electric power steel structure, and the system comprises a space-time sparse sampling module which analyzes the spatial distribution density of monitoring points of the electric power steel structure and the time domain characteristics of vibration signals, generates a dynamic sampling strategy, and carries out the acquisition operation through distributed sensing nodes; and the data transmission scheduling module encapsulates data according to a Predict Steel Core pre-research platform protocol, and performs directional transmission by adopting a time division multiplexing mechanism. The physical information prediction module imports an explicit time-domain physical information dynamic response prediction model and generates state prediction data in combination with the structure parameters; and the threshold early-warning analysis module loads an online incremental learning dynamic threshold early-warning algorithm to carry out real-time comparison. The platform interaction storage module stores data in a classified mode and provides a calling interface, the control coordination module dynamically adjusts operation parameters and data circulation time sequences of all the modules, efficient operation of the whole link is guaranteed, and the processing efficiency and accuracy of the safety production monitoring data of the electric power steel structure are improved.

The invention discloses an intelligent management and control system and method for the whole process of deep vertical shaft construction based on deep learning, relates to the crossing field of mine engineering and artificial intelligence, and aims to solve the problems of difficulty in complex spatial-temporal feature extraction, scarcity of labeled samples and weak dynamic working condition adaptation in deep vertical shaft construction monitoring. The system comprises three core modules of spatio-temporal feature extraction, collaborative enhancement semi-supervised optimization and reinforcement learning dynamic decision, firstly, multi-source heterogeneous data is collected through a sensor, deep spatio-temporal fusion features are extracted through an improved spatio-temporal convolutional network, and then a model is optimized by combining a small amount of annotated data with a large amount of unannotated data through a semi-supervised framework. According to the system, the monitoring accuracy is improved, the sample dependence and cost are reduced, the dynamic adaptability is enhanced, the intelligent management and control of the whole construction process are realized, the safety is guaranteed, and the efficiency is improved.

The invention discloses an edge reasoning optimization method and system based on segmented knowledge distillation. Firstly, a segmented knowledge distillation framework is constructed, a teacher-student model is divided into corresponding sub-modules with balanced parameters, parallel distillation training is adopted, middle feature dimension reduction and space alignment are achieved in combination with a principal component analysis method, the knowledge transmission efficiency is improved, and convergence is accelerated. Secondly, proposing an equipment perception self-adaptive pruning strategy, dynamically distributing a differential pruning proportion according to the real-time calculation capability and resource state of heterogeneous edge equipment, and balancing the load of low-performance equipment and the precision of high-performance equipment; and finally, establishing a deep reinforcement learning dynamic scheduling mechanism, generating a module delay-energy consumption file through offline analysis, adaptively selecting a device combination by an intelligent agent in an online stage, determining an optimal partition and deployment strategy through a threshold value distribution algorithm, and realizing joint optimization of energy consumption and reasoning time while meeting delay constraint.

The invention relates to the field of data processing, and provides a multi-dimensional feature driven B2B2C collaborative recommendation method and system. The method comprises the steps of performing multi-dimensional collection on B-end merchant features, C-end user features and commodity features through a heterogeneous data source interface to obtain standardized multi-dimensional features; performing dynamic weight learning on the standardized multi-dimensional feature data set through a multi-head self-attention mechanism to obtain a fusion feature vector; performing three-layer cooperative matrix construction on the fusion feature vector based on tensor decomposition to obtain a multi-dimensional factor matrix; performing causal relationship modeling on the multi-dimensional factor matrix through a causal graph structure-based collaborative filtering algorithm to obtain a deep collaborative network model; and performing real-time recommendation of to-be-recommended items through the deep collaborative network model to obtain a personalized B2B2C recommendation list. According to the method, the complex mode in the business scene can be captured, and the accuracy of the personalized recommendation result is improved.

The invention discloses an automatic voice response fault diagnosis method and system based on multi-source information fusion. The method comprises the following steps: receiving user voice input, synchronously obtaining user side intelligent electric meter data, meteorological environment information and historical service records, and extracting multi-modal features; semantic association is realized through an electric power service knowledge graph, and voice features, equipment data and environmental parameters are fused through space-time alignment; a machine learning dynamic decision tree engine is combined with a power consumption behavior analysis model to form a fault reasoning model, and a diagnosis result is effectively verified; and outputting the grading disposal scheme and triggering intelligent chemical order distribution. According to the invention, various information sources of the calling platform are fully utilized, a fault reasoning model with high accuracy is formed, the technical problems of inaccurate fault positioning and insufficient multi-source data collaboration of traditional voice response in power customer service are effectively solved, the diagnosis accuracy of customer power consumption problems is effectively improved, and the average processing time is shortened.

The invention discloses a fault monitoring and self-adaptive regulation and control method and system for a ship desulfurization system, and the method comprises the steps: collecting multi-source heterogeneous data in real time through a distributed sensor network, and carrying out the preprocessing; extracting multi-dimensional statistical features, performing weak fault feature enhancement by using an improved singular spectrum analysis algorithm, training a multi-channel attention mechanism LSTM network by using a high-discrimination feature sequence, learning a dynamic time sequence rule of a normal mode and a plurality of early fault modes, and performing early probability prediction of faults; identifying a complex fault source caused by coupling of a plurality of potential factors; a detection threshold value is dynamically adjusted by using an EWMA model driven by reinforcement learning, and the sensitivity and specificity of fault monitoring are dynamically optimized; and a fuzzy logic model is used for calculating a severity index, evaluating the severity of a fault, generating hierarchical early warning and iteratively optimizing a self-adaptive regulation and control strategy, so that the robustness and reliability of the system under strong noise and multivariable coupling are improved.

The invention provides a vehicle network interaction AI cooperative scheduling method and system based on multi-source data fusion. According to the method, a'federated learning node + abnormal data processing unit 'collaborative architecture and an AI collaborative scheduling model containing topological modeling, prediction and optimization decision submodules are innovatively constructed, and a unique technical system is formed; fusing vehicle, power grid, environment, user and charging station multi-source heterogeneous data, and generating a vehicle-power grid topological association feature set through isolated forest + GAN preprocessing and transverse federated learning encryption fusion; gAT heterogeneous graph modeling, LSTM-Transform prediction, deep reinforcement learning optimization and online learning dynamic parameter adjustment technologies are integrated, and the data processing efficiency is improved; the data privacy security (the original data cross-domain transmission quantity is equal to 0) is realized, the topological association precision is improved by 40%, and the manual risk and the management cost are reduced; the application module generates a charging and discharging instruction, a load adjusting quantity and an excitation scheme, the scientific management of the whole process of vehicle network interaction is supported, the peak-valley cost is reduced by 15-20%, and the carbon emission is reduced by 10-15%.

The invention discloses an accelerator architecture design method and electronic equipment, relates to the technical field of computers, and performs dynamic fractal space coding on a design parameter matrix of an accelerator to obtain an analog coding vector. And performing meta-learning dynamic weight entropy search on the analog coding vector by using a historical task data set to determine a new design scheme. And when the difference between the sensitivity parameters in the new design scheme and the sensitivity parameters in the neighbor set meets a difference condition, performing multi-objective optimization simulation on the new design scheme. And when the difference does not meet the difference condition, multiplexing the simulation result of the neighbor set. And performing iterative search according to Bayesian optimization, and determining a final accelerator architecture design scheme in the selected design scheme and the simulation result thereof. Dimension reduction is performed on a high-dimensional discrete space, and a search direction is guided by means of an efficient meta-learning dynamic weight entropy acquisition function. And rapid exploration of an accelerator architecture design space under multi-objective optimization is realized.

The invention discloses an intelligent power grid dispatching system and method based on digital twinning and deep learning, and relates to the technical field of intelligent power grid dispatching, the intelligent power grid dispatching method based on digital twinning and deep learning specifically comprises the following steps: step 1, multi-source heterogeneous data and a dynamic parameter updating mechanism are fused, and a dynamic parameter updating mechanism is established; the method comprises the following steps of: 1, constructing a power grid digital twinborn body synchronously mapped with a physical power grid, 2, aggregating multi-region power grid data by adopting a federated learning framework, establishing an equipment degradation model and perfecting a priority data channel mechanism, and 3, determining a real-time power grid state and an equipment risk coefficient based on the power grid digital twinborn body, and designing a deep reinforcement learning dynamic decision framework. According to the method, the real-time power grid state and the equipment risk coefficient are determined based on the power grid digital twinborn body, a deep reinforcement learning dynamic decision framework is designed, and multi-target collaborative load scheduling optimization can be achieved.

The invention provides a reinforcement learning dynamic pricing method based on game feedback, and relates to the technical field of online data market dynamic pricing. Through combination of leader-follower strategy interaction of the Stackelberg game and the exploration-utilization balance principle of the dobby machine, the problems of price adjustment lag, large income fluctuation and insufficient strategy stability in a non-stable environment are solved. By introducing a data freshness evaluation mechanism, data weight is dynamically adjusted by integrating data timeliness attenuation and market correlation analysis, it is ensured that data input into a model always has high real-time performance and strong correlation, and the problem of decision lag caused by static data input in a comparison scheme is avoided; the exploration-utilization balance principle and the reinforcement learning model of the dobby machine are fused, the exploration rate is dynamically adjusted, a historical optimal strategy can be fully utilized in a non-stationary environment, a potential better scheme can be explored, and the static optimization bottleneck that a game model is limited to a preset objective function in a comparison scheme is broken through.

The invention discloses a radar interference effect evaluation method based on a constraint learning dynamic Bayesian network, is applied to the field of radar interference evaluation, and aims at solving the problem that the accuracy of interference effect evaluation is reduced due to radar detection data missing in a complex electromagnetic environment. Meanwhile, parameter constraints of five types of evaluation indexes and interference effect grades are defined; secondly, constructing a constraint learning dynamic Bayesian network, and learning a conditional probability and a transition probability under a data missing condition; then, proposing a prior constraint expectation maximization algorithm, converting parameter learning into an optimization problem with constraint by combining convex optimization, and overcoming the defects of a traditional expectation maximization algorithm; secondly, a cloud model is introduced to quantify discrete probability distribution into a continuous interference degree value; finally, simulation shows that the method can effectively improve parameter learning stability and evaluation accuracy under the conditions of suppressing and deception jamming and single index deficiency, and provides a reliable scheme for radar jamming effect evaluation in a complex environment.

The invention discloses a power grid supply chain supplier risk intelligent assessment system, which realizes full-dimension assessment of supplier risks through dynamic index construction, multi-model collaborative analysis and adaptive weight adjustment technologies. The system integrates structured data, a graph network relationship and unstructured text information, and combines reinforcement learning to dynamically optimize the weight, so that the real-time performance and accuracy of risk assessment are remarkably improved.

The invention discloses a generalized energy storage equivalence model construction method based on multi-type micro-grid mapping, and relates to the technical field of energy storage model construction, and the method comprises the steps: dividing an energy storage unit into a plurality of equivalence model candidate structures through a graph neural network model based on an electrical topological graph and a node state of an energy storage system; model parameters are quickly initialized and calibrated online based on meta learning; deep reinforcement learning dynamic decision structure switching and parameter optimization are introduced; and adopting Monte Carlo Dropout to evaluate the parameter confidence and trigger model rollback, and outputting a final energy storage equivalent model. According to the method, the accuracy and robustness of the model under the dynamic working condition are improved, the uncertainty in the modeling process can be sensed in real time, rollback correction is actively carried out, the reliability and engineering availability of model output are ensured, and the method is particularly suitable for energy storage system modeling and control scenes with high requirements for the response speed, the adaptive capacity and the operation stability.

The invention discloses a federated learning dynamic aggregation method, system and device, a medium and a product. According to the method, a twinborn layer is constructed on a central server, model aggregation deduction is performed in advance in a virtual environment by establishing a digital twinborn model of a client, and then an optimal aggregation strategy is screened out; in each round of iteration, the client maps self state information and local model parameters to a twinborn layer, the twinborn layer clusters the client by using a K-means dynamic clustering algorithm, the aggregation effect of different cluster combinations is evaluated, and an optimal aggregation strategy is selected for pre-aggregation; therefore, the server globally aggregates the client model according to the strategy provided by the twinborn layer. According to the embodiment of the invention, the model convergence speed and precision of federal learning can be remarkably improved, the communication delay and energy consumption are reduced, the method effectively adapts to the isomerism of clients, and an efficient and flexible solution is provided for distributed machine learning.

The invention relates to the technical field of production scheduling and intelligent manufacturing, and provides an assembly line scheduling optimization method considering generalized priority constraints, which comprises the following steps: constructing an integer linear programming model, performing generalized priority relation enhancement and propagation on an input instance of the integer linear programming model, and improving the capacity of a workstation; calculating a simple lower bound based on task processing time, a lower bound based on the longest path of the priority graph and a lower bound based on a maximum flow algorithm, and taking the maximum value of the three as a final lower bound; defining an action space, defining a state space to represent the change of a current solution on an objective function value and a distribution balance degree relative to a global optimal solution and a local optimal solution, and designing a reward function to adaptively select a neighborhood operation; constructing an initial scheme based on the final lower bound and a Q-learning dynamic selection module, and performing batch movement iteration through neighborhood operation adaptively selected by the Q-learning dynamic selection module; and outputting an assembly line scheduling scheme meeting the generalized priority constraint based on a result of the batch movement iteration local search.

The invention provides an FCBGA chip support plate process parameter optimization method, system and device and a medium, and belongs to the technical field of PCB production, the method comprises the following steps: collecting and preprocessing a historical data set in the FCBGA chip support plate production process, and screening data subsets with correlation greater than a preset threshold based on target features of a to-be-optimized production task; training a machine learning prediction model capable of mapping the process parameters to the quality indexes; constructing a multi-objective optimization function by taking a machine learning prediction model as a core; a multi-objective optimization algorithm is adopted to search in the process parameter constraint space, and a Pareto optimal solution set is solved; and selecting a final optimal process parameter according to a preset decision strategy, applying the final optimal process parameter to an actual production line of the FCBGA chip carrier plate, then collecting actual production result data, and feeding back the actual production result data to a historical data set. The method is based on historical data machine learning multi-objective optimization, incremental learning dynamically adapts to production line changes, and quality and efficiency are improved.

A processor may capture network packets from a communication channel between a user and an application, the network packets comprising a data structure package. A processor may identify a schema of a payload used in the data structure package of the network packets. A processor may dynamically generate payload schema processing code using a machine learning dynamic protocol parser by applying a machine learning model to the schema of the payload, the payload schema processing code including a description of each field of the data structure package and a parser function for extracting a prompt. A processor may iterate the dynamically generating of the payload schema processing code until the payload schema processing code meets a predefined accuracy and functionality criteria that successfully extracts the prompt.

The invention discloses a non-intrusive load identification method based on reinforcement learning dynamic simulation, and relates to the technical field of power consumer side energy consumption monitoring. The method comprises the following steps: firstly, collecting time sequence data of various types of electric equipment in independent operation and a plurality of superimposed operation scenes; then constructing and training an electrical coupling data conversion model used for learning an electrical characteristic mapping rule between single-device independent operation and multi-device superposition operation; meanwhile, a scene generation strategy network is constructed and trained to automatically generate a diversified multi-device superposition operation scene; then, a massive simulation training sample training classification model close to a real scene is generated through cooperation of the trained conversion model and the strategy network; according to the method, the problems of training data shortage and feature mismatch caused by infinity of equipment start-stop time sequence combination and an electrical coupling effect are effectively solved, and the precision, robustness and generalization ability of a non-intrusive load identification model in an actual complex power utilization environment are improved.

The invention discloses a credit risk joint modeling system and method based on federal learning. The system comprises a data preprocessing module, a heterogeneous data adaptation module, a federal learning dynamic modeling module, a security communication module and a risk assessment decision module. The data preprocessing module adopts a hierarchical encryption strategy, sensitive fields are subjected to CKKS homomorphic encryption, super-sensitive fields are bound by adopting TFHE full homomorphic encryption and combining with biological characteristics, and non-sensitive fields are encrypted by adopting SHA-3 Hash. And the heterogeneous data adaptation module realizes cross-mechanism data semantic alignment through a three-level mapping network and adversarial training. And the federated learning dynamic modeling module is used for carrying out local training by using a modified MobileNet-V3 network, and dynamically adjusting an aggregation weight based on Bayesian optimization. And the secure communication module ensures data interaction security based on the block chain and zero-knowledge proof. According to the method, multi-mechanism collaborative modeling without local data is realized, the AUC value of the model is increased from 0.82 to 0.94, and the accuracy of credit risk assessment is effectively improved.

The invention provides an airport terminal building heating, ventilation and air conditioning load prediction system and method based on a micro-service architecture. The method comprises the following steps: collecting multi-source heterogeneous original data and extracting a key feature set; mining space-time relevance between people flow and equipment operation, and outputting a space-time characteristic matrix; and constructing a multi-algorithm fusion framework, carrying out cross-terminal distributed cooperative training and dynamic weight aggregation through federal learning, introducing a lion group optimization algorithm to optimize global parameters, and outputting an airport terminal heating ventilation air conditioning load prediction result. Fusion data are standardized, key features are extracted in combination with automatic feature engineering, and the data utilization rate and feature relevance are improved; then capturing a space-time association mode by using a space-time diagram convolutional network, constructing a multi-algorithm fusion framework and dynamically allocating weights; then realizing distributed cooperative training by adopting federal learning, and optimizing parameters in combination with a lion group optimization algorithm to enhance the generalization ability of the model; and finally, through incremental learning dynamic iterative optimization, the prediction precision is continuously improved.

The invention discloses a multi-modal sensing and edge intelligent decision-making self-adaptive feeding system and method, and relates to the technical field of intelligent logistics equipment, industrial automation and information physical systems, a visual system module is combined with distance measurement, weighing and recognition detection units of a sensor module, multi-modal sensing of the material state and environment is achieved, and intelligent edge decision making is achieved. The edge decision module dynamically optimizes feeding parameters based on reinforcement learning, the edge prediction maintenance module guarantees equipment health, and the system execution module achieves accurate control and cleaning. The feeding method comprises the steps of multi-modal data synchronous acquisition, digital twin model real-time updating, AI multi-target optimization decision making, multi-actuator cooperative action, industrial Internet of Things cooperative interaction and feedback-based iterative optimization, and multi-modal sensing, edge intelligent decision making and accurate execution are fused. Self-sensing, self-decision-making, self-execution and self-optimization in the feeding process are achieved, the feeding efficiency and quality are improved, and accidental shutdown and comprehensive cost are reduced.

The invention discloses a cross-device federated learning method and device, equipment and a medium, and relates to the technical field of cross-device federated learning incentive, and the method comprises the steps: firstly obtaining a weight matrix representing the data distribution condition of each edge client, and a feature vector of the weight matrix, and obtaining a norm of the feature vector corresponding to each type of elements in the weight matrix; according to the method, data distribution of each edge client is deduced based on norms, enhanced data is designed for each edge client according to a data missing proportion, in the process, client data distribution is accurately deduced through analysis model updating, and enhanced data for a data missing category of each client is generated for each client; meanwhile, a deep reinforcement learning dynamic multi-round auction strategy is combined, an incentive strategy is made, whether enhanced data is used to participate in training or not is selected, the method does not depend on local data on the whole, a distributed incentive decision-making mechanism abandons a traditional centralized decision-making framework, efficient and intelligent distribution of incentive resources is achieved, and the training efficiency is improved. Therefore, the overall performance of cross-device federated learning is greatly improved.