50 results about "Global strategy" patented technology

The invention relates to the technical field of industrial production lines, and discloses an industrial production line multi-device dynamic collaborative scheduling method based on reinforcement learning, comprising the following steps: S1, modeling a three-dimensional state space; s2, hierarchical reinforcement learning architecture; and S3, edge-cloud cooperative execution. According to the industrial production line multi-device dynamic collaborative scheduling method and system based on reinforcement learning, device states, task constraints and resource occupation are integrated into a structured matrix through three-dimensional state space modeling, and a global decision-making layer captures production time sequence dependence by using a bidirectional long-short-term memory network; modeling equipment space association and process constraints through a graph attention network, and generating a global strategy including task allocation, capacity adjustment and resource pre-allocation; and after the edge layer detects the dynamic event, the cloud platform generates a candidate scheme through Monte Carlo tree search, and realizes dynamic event response and multi-target collaborative optimization by combining multiple targets such as global value network evaluation task completion time and equipment load balancing.

The invention provides a distributed collaboration-based computing network intelligent agent system and a resource dynamic scheduling method thereof, and belongs to the technical field of computing network integration. The system comprises an edge agent used for sensing local computing power, network bandwidth and task load in real time, predicting task demand fluctuation by using a lightweight neural network, adjusting resource allocation weight in real time in combination with network topology change, and executing a preliminary task scheduling decision; the regional collaborative agent is used for aggregating multiple edge node states based on federated learning, generating a cross-node resource scheduling strategy, verifying the credibility of a computing power transaction smart contract and determining a cross-domain resource allocation scheme; and the cloud management agent is used for constructing a global resource portrait model according to the information provided by the edge agent and the regional collaborative agent, performing long-term strategy optimization, issuing global strategy information, and constructing and updating a computing power transaction smart contract based on a preset computing power transaction smart contract template. According to the invention, multi-level refined scheduling of computing network resources is realized.

The invention relates to the technical field of intelligent dialogues, and discloses a dialogue marketing strategy optimization method based on artificial intelligence, and the method comprises the following steps: collecting the voice, text, facial expression and physiological signals of a user in real time through a multi-modal perception assembly of a terminal device, constructing a dynamic emotion map, and extracting a multi-modal feature vector; a cross-modal information processing module is utilized to align the multi-modal data through a comparative learning algorithm, and causal relationship description of user behaviors and strategies and strategy risk scores are generated; cooperatively training a global strategy model through differential privacy and homomorphic encryption technologies; combining the dynamic emotion map and a causal model to generate an emotion adaptive dialogue script, and optimizing strategy selection through a reinforcement learning algorithm; and dynamically updating the emotion map, the risk score and the global model through a closed-loop feedback mechanism to form a real-time optimized strategy generation system. According to the invention, the practicability of artificial intelligence to real-time services can be improved.

The invention discloses a data processing system and method based on a federated architecture interface and related equipment, and relates to the technical field of computer software, in particular to a federated architecture distributed system formed based on a central coordination node and a plurality of regional autonomous nodes. The central coordination node is used for performing global strategy making, metadata block chain evidence storage, conflict detection and federated monitoring and self-healing to realize global management and control, and the regional autonomous node adapts to a physical machine heterogeneous environment to perform local strategy execution, heterogeneous protocol conversion and resource autonomous regulation and control so as to realize flexible regulation of a regional strategy. According to the scheme, through a federated and intelligent treatment system, inherent contradictions between centralized management and control and department autonomy, standardization and rapid iteration and global safety and local flexibility are solved, technical support is provided for an enterprise to construct a standardized, high-reliability and easy-to-expand API ecological system, digital transformation is assisted, and the enterprise safety is improved. And global controllability and local autonomy of enterprise-level API ecology are realized.

The invention relates to the technical field of telecommunication. By providing the multi-domain strategy collaborative dynamic decision optimization method, system, equipment and medium, the method comprises the following steps: extracting cross-domain common characteristics of network layer topology, equipment state and alarm information, retaining differential characteristics of a transmission network and a data network, and generating a uniform resource data set; on the basis of the unified data set, a global optimization strategy is generated through linkage decision making of a three-level collaboration module; matching the strategy into an atomic instruction set executable by a heterogeneous network manager by utilizing semantic ontology mapping and syntax tree matching; and according to the instruction execution feedback data, dynamically adjusting the strategy weight and carrying out iterative optimization, and generating an updated global strategy, so as to achieve the technical effects of improving the cross-domain resource perception precision, enhancing the hierarchical decision collaborative efficiency and reducing the multi-manufacturer instruction conversion error rate.

The invention relates to the technical field of artificial intelligence service development and intelligent operation and maintenance, and discloses a system for developing and designing an AI service, and the system comprises a multi-dimensional target configuration module which defines a business target and a KPI, and generates an initial configuration parameter; the life cycle layered modeling module is used for receiving parameters and constructing a digital twinborn model in a layered (macroscopic, mesoscopic and microscopic) manner; the global prediction optimization module is used for executing multi-objective optimization based on digital twinning and generating a global control strategy; the distributed intelligent execution module is used for executing a global strategy in a microcosmic layer by an intelligent agent and performing local autonomous optimization; the intelligent resource management module is used for dynamically distributing system resources according to strategies and task requirements; and the monitoring module collects execution and resource data and is used for updating the digital twinborn model and adjusting a global strategy. According to the method, AI service accurate modeling, global intelligent optimization, efficient cooperative execution and continuous closed-loop self-adaption are realized.

The invention discloses a data center intelligent energy-saving control system and method based on artificial intelligence, and the system comprises an SLA-energy-saving collaborative decision center, a business support module and a feedback iteration module, and the SLA-energy-saving collaborative decision center is in communication connection with the business support module and the feedback iteration module. An intelligent energy-saving coherent process of demand analysis, strategy generation, execution feedback and iterative optimization is formed, and action linkage of all links is achieved through real-time data interaction; a unified data model unit, an SLA constraint disassembly unit and a global strategy generation unit are arranged in the SLA-energy-saving collaborative decision center, reliability requirements of different tenants are converted into specific control standards through the SLA constraint disassembly unit, and the reliability requirements of the different tenants are optimized by combining a priority mechanism of task scheduling and differential configuration of all equipment parameters. The SLA standard reaching rate of 99.999% or above of the core tenants can be guaranteed, and more energy-saving space can be mined for the common tenants and the basic tenants.

The invention discloses a smart city-oriented cloud resource adaptive scheduling method and system, and particularly relates to the technical field of cloud resource adaptive scheduling, and the method comprises the steps: precisely recognizing a high-risk task through calculating a task mobility misjudgment coefficient, and marking the high-risk task for deep risk control; calculating a node resource pseudo available coefficient for candidate nodes to be migrated by the high-risk task; a strategy degradation risk analysis model is constructed based on the coefficients to quantify and output a global strategy degradation trend, a'task misplacement-resource misplacement 'closed-loop fault mode is effectively revealed, and a quantifiable basis is provided for risk early warning and strategy correction; and on the basis of the evaluated high degradation risk degree, constructing a scheduling objective function to realize real-time optimization of task-node matching.

The invention relates to a cloud edge multi-unmanned aerial vehicle cooperative resource optimization method assisted by a large language model, and belongs to the technical field of unmanned aerial vehicle communication, and the method comprises the following steps: S1, constructing an edge-cloud unmanned aerial vehicle cooperative reasoning system; s2, establishing a joint optimization model for discriminating gain maximization; s3, deploying a large language model at a cloud node, and generating a global strategy through a planner, a memory bank and an reflection evaluator; s4, deploying a deep reinforcement learning model at each edge node, and executing real-time optimization according to a global strategy and local observation data; s5, a collaborative feedback mechanism is established, the edge node feeds back an execution result to the cloud node, the cloud node updates a global strategy according to the feedback result, and the edge node adjusts real-time optimization parameters according to the updated global strategy; and S6, adopting an actor-commentator resource allocation algorithm for dynamic knowledge flow collaborative optimization, and realizing collaborative optimization through a distributed sensing and centralized decision framework.

The application discloses a kind of microservice dynamic access control strategy under zero trust architecture method and device, belong to network security, distributed system, zero trust, microservice field, it specifically includes: by the event tracker of node to obtain the context information in microservice runtime context event, context manager collects and aggregates context information and is reported to microservice control face global manager, global manager is according to global context mapping table value, judge whether it is new microservice instance, if yes then update global and local Tag mapping table, when node receives message, message verifier obtains the Tag value carried by this message, by local Tag mapping table to obtain the microservice context information corresponding to this message, after inquiring local strategy table and global strategy table, run decision algorithm, execute corresponding access control strategy, use context to identify communication endpoint, with dynamic, accurate, scalable advantage.

The invention belongs to the field of multi-satellite cooperative task planning of deep learning, and particularly relates to a multi-satellite cooperative task planning method based on deep learning, which comprises the following steps: constructing a global strategy skeleton; constructing a local immediate adjustment strategy, specifically, constructing a constraint perception greedy scoring formula; constructing a feasibility projection layer formula; constructing a preferential and locking formula; according to the method, rapid low-delay recalculation and acquisition and transmission consistency improvement can be realized, so that the completion rate is improved, rollback and jitter are reduced, and the overall benefit and service fairness are improved on the premise of guaranteeing resource security.

The application relates to a multi-agent reinforcement learning training method and system, which comprises the following steps: a local strategy training stage, in which an agent makes an action by using local observation information; a global strategy training stage, in which a global strategy of the agent uses the local strategy as an action module for interacting with an environment, the global strategy takes global state of the environment as input, encodes the global information in a hidden space, and uses a neural network to fit the global state by using local observation information of all agents; the local strategy makes a suitable action in the environment according to the local observation information and the output of the global strategy; and a local strategy optimization stage, in which the global strategy, the local strategy and a fitting model obtained in the previous two stages are used to optimize the existing local strategy, so that an agent with better effect is finally obtained. The application can improve the speed and accuracy of multi-agent reinforcement learning training.

The application discloses a kind of family energy storage collaborative optimization control method, device, equipment and storage medium.The method comprises: obtaining the operating state data and target income expectation of family energy storage network;Operating state data and target income expectation are input into family energy storage collaborative control model, and target collaborative control strategy is determined;Wherein, family energy storage collaborative control model includes: characteristic aggregation module, causal coding module, action mapping module and projection constraint module, characteristic aggregation module is used to carry out graph-level feature aggregation, determines global state vector and the node embedding feature of each energy node;Causal coding module is used to carry out causal strategy intention coding, determines global strategy intention vector;Action mapping module is used to determine the action scalar of each energy node;Projection constraint module is used to carry out power mapping, and determines target collaborative control strategy.The application can realize flexible and efficient control to family energy storage, guarantee the economy and robustness of energy storage control.

The invention discloses a big data security access control method. The method comprises the following steps: collecting multi-dimensional information of an access request, including time, place, equipment, network environment and historical behaviors, and generating a risk score in combination with a weight; and converting the risk score into a risk probability by using a probability mapping model, and dividing risk levels according to a preset threshold. And the system forms a dynamically adjustable permission set according to the basic permission of the user role and the limit permission corresponding to the risk level, and calculates a security strength index. And further combining the risk score with the security strength to generate a comprehensive risk coefficient and recording the comprehensive risk coefficient in an audit log. And when the risk coefficient exceeds a threshold value and exceeds the limit continuously for multiple times, triggering a security response mechanism, including account freezing, session blocking, alarm pushing and global strategy tightening. According to the method, real-time evaluation, fine-grained control and linkage response of big data access are realized, and the dynamic defense capability and the overall security of the system are improved.

The application provides a method for simultaneous localization and mapping based on hierarchical reinforcement learning, comprising: obtaining a hierarchical reinforcement learning model, which comprises a localization module, a map reconstruction module, a global strategy module, a planner and a local strategy module; continuously obtaining a pair of observation maps, and determining a control action for moving the robot in the environment, a localization result and an updated map corresponding to the environment based on reinforcement learning by the hierarchical reinforcement learning model, wherein training data is collected after the current control action is performed for the reinforcement learning of the global strategy module, each piece of training data comprising a global reward corresponding to a control action, the global reward being positively correlated with exploration efficiency and map accuracy and additionally rewarding exploration efficiency when the control action leads to a loop signal.

The application belongs to the technical field of trusted cloud, and provides an automatic and hierarchical trusted cloud level management unit system and method, which comprises the following steps: a robust autonomous unit TCDUN is constructed for continuous monitoring, and whether to trigger the dynamic level recalculation of the TCDUN is judged according to the monitoring result, and the dynamic level recalculation is executed when triggered, and a trusted state report is generated to respond to a remote proof request; a security dimension requirement is extracted according to the input, and the level requirement mapping of three dimensions is carried out, a product combination meeting the requirement is found according to the mapping result, and the weight of a product with quantum-resistant cipher characteristics is adaptively optimized; a global strategy is issued by a superior TCDMU, a subordinate TCDMU receives and penetrates to a TCDSU strategy, and an execution state reporting and compensation mechanism is realized, so that a complete automatic management cycle is realized, and the whole-process automatic management and control of the trusted cloud unit is achieved.

The application discloses an automatic driving centralized decision-making method based on a hybrid layered reinforcement learning, comprising the following steps: constructing a deep reinforcement learning network and performing hybrid network layering on the deep reinforcement learning network to obtain an upper network and a lower network; performing horizontal division on the upper network to obtain a horizontal global strategy network and a vertical global strategy network; training the horizontal global strategy network and the vertical global strategy network through a discrete DRL algorithm, wherein the horizontal global strategy network outputs a lane-changing instruction, and the vertical global strategy network outputs a vertical control strategy; dividing the lower network into a vertical lower network and a horizontal lower control network, training the vertical lower network and the horizontal lower control network through a continuous DRL algorithm, wherein the vertical lower network selects and activates a corresponding sub-strategy network according to the network output of the vertical global strategy and outputs continuous acceleration; and the horizontal lower control network changes lanes based on an optimal control rule to complete centralized decision-making of automatic driving. The application further improves the accuracy, comprehensiveness and effectiveness of decision-making.

The application discloses a kind of heterogeneous unmanned aerial vehicle cluster efficient cooperative task planning method, belong to unmanned aerial vehicle cluster task planning field, including the following steps: S1: input heterogeneous cluster unmanned aerial vehicle total number, opposite ground target number, role division interval time step, constraint strength between global strategy and role strategy, predefined role number;S2: by the representation of each action in action space and the representation of each unmanned aerial vehicle state, realize the role division of unmanned aerial vehicle, i.e. intelligent agent;S3: according to the state observation of unmanned aerial vehicle, dynamically carry out unmanned aerial vehicle role selection, update every T time step;S4: trajectory data is generated and stored by role strategy sampling, and the decision strategy of global cooperative planning is obtained by initialization training;S5: according to role strategy, constantly update trajectory data and train guide global strategy, finally obtain global strategy and carry out model inference.

The invention relates to the technical field of service evaluation, in particular to an ecological service evaluation method and device for a marine ranch, a medium and a product. The method comprises the following steps: acquiring historical production data and current production data of a historical production cycle of a marine ranch, wherein the historical production data and the current production data both comprise data of operation type indexes; generating an ecological service evaluation result of the current production cycle based on the historical production data and the current production data; formulating a global strategy for the next production cycle of the operation type index; and executing the next production cycle based on the global strategy, and dynamically adjusting the global strategy based on the production data of the next production cycle. The production strategy is dynamically adjusted based on the ecological service evaluation result, and balance of ecology and production is achieved.

The application discloses a kind of federal reinforcement learning method and system based on hessian auxiliary and strategy gradient under heterogeneous environment, including server and multiple clients, server initializes global model parameter, global gradient update quantity and global control variable and broadcasts to client, client takes current global model parameter as local update starting point, obtains main trajectory and hessian estimation auxiliary trajectory by parameter interpolation and double trajectory sampling, constructs hessian auxiliary correction term at interpolation parameter, and forms local gradient estimation in combination with control variable correction term, uploads parameter update quantity and control variable update quantity after completing multi-step local update, server aggregates each client upload result, updates global model parameter and global control variable, and outputs global strategy model parameter after cyclic iteration, completes federal reinforcement learning.The application can weaken the client drift caused by heterogeneous environment, reduce strategy gradient estimation fluctuation, improve training stability and global strategy performance.

The present invention discloses a federated reinforcement learning method for autonomous driving based on meta-learning and experience vectors, comprising: constructing a heterogeneous environment for autonomous driving; constructing an Actor network, and updating the Actor network parameters through deterministic policy gradients; constructing a Critic network, and updating the Critic network parameters through a meta-learning algorithm and an error back-propagation algorithm. During the training process, experience vectors are added to integrate and update the learning experience of each autonomous driving vehicle to optimize the performance of the global strategy; during the training process, all autonomous driving vehicles regularly send local Critic network parameters, Actor network parameters, and local experience vectors to a central server for aggregation. The present invention solves the problems of slow convergence and decreased strategy generalization ability in existing federated reinforcement learning due to dynamic changes in the environment and fluctuations in strategy performance by improving the adaptability of autonomous driving vehicles to environmental heterogeneity, thereby improving training efficiency and strategy robustness.

The application discloses a safe and credible host access risk control system and method based on a centralized and distributed control, and relates to the technical field of network security guarantee.The application comprises the following steps: a central controller performs host verification based on a multi-dimensional verification system of credible verification; at least one host controller is arranged at a subnet entrance as a dispersedly executed edge control node; a control object layer is formed by various hosts; the risk control system adopts a centralized and distributed control architecture, and parallel computing engines constructed in the local host controller are used to split and process the control tasks in parallel; the central control node is responsible for global strategy formulation, resource scheduling and high-risk event research and judgment, and the uniformity of the control standard is guaranteed; the host controller is responsible for local access request preprocessing, basic risk assessment and low-risk event disposal, and realizes nearby response; all host devices in the network are controlled in a centralized manner based on the edge security execution component, and the application is suitable for the security control of business terminal computing environments.

The invention provides a vehicle-network interaction power-carbon coordinated regulation and control method and related equipment. The method comprises the following steps: acquiring vehicle network data, and performing preprocessing and time-space alignment on the vehicle network data to obtain a time-space unified data set; constructing a physical execution graph based on the space-time unified data set; obtaining a predefined global strategy graph; based on a preset node mapping function, mapping each node in the global strategy graph to a corresponding node in the physical execution graph, and generating a state variable and a control variable of the corresponding node in the physical execution graph based on the state variable and the control variable of each node in the global strategy graph; performing joint prediction based on the physical execution graph and the time-space unified data set, and inputting a prediction result into a node state of the global strategy graph; generating a regulation and control strategy set based on the global strategy graph after node state updating; and generating and issuing a regulation instruction based on the regulation strategy set.

This invention discloses a human energy collaborative management system and method based on digital twins and multi-agent reinforcement learning, belonging to the field of artificial intelligence and precision health management technology. The system includes a multimodal data acquisition module, a human energy digital twin modeling engine, an energy production optimization agent, an energy transfer regulation agent, a consumption reduction management agent, a collaborative reinforcement learning decision engine, and a user interaction module. The method constructs an individualized digital twin model to represent the state of the human body's energy production, energy transfer, and consumption reduction in real time; generates preliminary strategies for each stage using three dedicated agents; optimizes and integrates global strategies through a centralized evaluation-distributed execution collaborative reinforcement learning architecture; and outputs personalized intervention plans after simulation verification and safety screening in the digital twin. This invention overcomes the problems of fragmented stages, static regulation, and insufficient personalization in existing health management technologies, achieving closed-loop, adaptive, and collaborative management of the entire human energy metabolism chain. It is particularly suitable for improving the energy balance efficiency and physiological resilience of the human body in extreme environments such as conventional and space environments.

The present invention discloses a cloud-edge collaborative robot cluster simulation training and optimization system, comprising: a node labeling unit, a node classification unit, a dual-model construction unit, a reinforcement learning unit, and a periodic parameter collection unit for periodically collecting real-time cloud model parameters and local model parameters; a parameter fusion unit for unifying the format and aligning the timestamps of cloud model parameters and local model parameters, and fusing them into global parameters; a global model application unit for constructing the global parameters into a global strategy model under a reinforcement learning framework for use in simulation optimization training of the target system; the present invention improves the clarity of computing resource configuration, realizes the unified expression of cross-node task objectives, provides an input basis for centralized strategy optimization, and improves the consistency of cloud-edge collaboration.

This invention relates to the field of intelligent building environment control technology, specifically an adaptive environmental regulation system. It utilizes an AI strategy module with an integrated edge computing platform, housed in the main control cabinet. Based on outdoor weather forecasts, local climate information, and group behavior patterns, the module performs multi-parameter environmental control strategy reasoning to generate adaptive environmental regulation commands. Simultaneously, a master-slave hierarchical control architecture is employed. Based on centralized optimization decisions in the main control cabinet, regional autonomous regulation and user interaction are achieved through various user terminals. The AI ​​strategy module further uses predictive models and optimization algorithms to predict future environmental loads based on historical environmental data and real-time conditions, calculating the optimal control parameters for each environmental regulation device. This system not only significantly improves the accuracy and foresight of environmental regulation through AI predictive control and real-time optimization, but also ensures global strategy coordination and energy efficiency optimization through the master-slave hierarchical architecture, guaranteeing rapid response and system reliability for regional control.

The invention relates to the technical field of new energy automobile charging facility management and digital twinning, and particularly discloses a new energy automobile charging demand prediction and scheduling management system, which comprises a global digital twinning body, a distributed prediction scheduling agent cluster, a regional collaborative arbiter and a global strategy coordination center. The proxy cluster carries out demand prediction and optimal scheduling based on local data; the regional arbiter coordinates agent decisions to prevent regional resource conflicts; and the global strategy center optimizes the system strategy through reinforcement learning. Through the architecture, localization, real-time and global coordination of charging scheduling decision are realized, and the response speed, fault tolerance and overall operation efficiency of the system are improved.

The application relates to a large language model autonomous evolution method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring current environment information and a global strategy; inputting the current environment information and the global strategy into a multi-model set; generating an action instruction based on the current environment information and the global strategy by each large language model, and generating an action instruction based on the current environment information and a mathematical model by each reinforcement learning algorithm; selecting an optimal instruction with the highest accuracy from all the action instructions output by the multi-model set, updating the global strategy based on the optimal instruction, and completing one evolution cycle of the global strategy; based on the updated global strategy and the current environment information, the multi-model set is used to generate the action instruction again, and the steps of selecting the optimal instruction and updating the global strategy are repeatedly executed until a preset cycle termination condition is reached. The application improves the efficiency and stability of the large language model evolution process.

The application belongs to the technical field of power system operation control, and relates to a three-phase distribution network voltage optimization control method based on vertical and horizontal flexible resource cooperation and large language model enhanced federal reinforcement learning. The method sets up a three-phase distribution network power flow model and defines a voltage optimization problem, sets a state space, an action space and an initial reward function of multi-agent reinforcement learning; each regional agent obtains a low-dimensional state vector by using a large language model and a data distillation algorithm, constructs a phase correlation matrix by using phase characteristics, and generates weighted features; each regional agent independently performs local reinforcement learning training, performs constraint solving based on the three-phase distribution network power flow model, generates a semantic abstract uploaded to a computing center, generates global strategy guidance information, and issues the global strategy guidance information to each agent; the trained agent is deployed to a distribution network control system, action instructions are generated according to real-time collected system states, and vertical and horizontal flexible resources are cooperatively controlled.

The invention discloses a thermal power plant equipment maintenance strategy optimization method and system based on reinforcement learning, and relates to the technical field of thermal power plant intelligent operation and maintenance. The method comprises the following steps: constructing a multi-agent system which takes optimization of a global strategy as a common target by a plurality of equipment components with a dependency relationship, and establishing a partially observable Markov decision process model which covers information such as residual service life and productivity for each agent, meanwhile, discrete action spaces such as non-maintenance, imperfect maintenance and replacement are set; a multi-agent deep reinforcement learning algorithm is applied, and a global maintenance strategy capable of maximizing a multi-target long-term expectation utility function is learned by interacting with an environment model and utilizing an experience playback technology for training; and outputting an optimal combination maintenance action for a real-time state according to a maturely trained strategy. According to the method, global coordination and dynamic self-adaptive maintenance optimization of the complex equipment of the thermal power plant can be realized, and the economical efficiency and the risk are effectively balanced.