636 results about "Value network" patented technology

The invention provides a multi-target collaborative optimization control method and system for an igniter robot production line, and relates to the technical field of automatic control, and the method comprises the steps: collecting real-time operation parameters of the production line, carrying out the wavelet denoising, feature extraction and fusion, inputting a hierarchical asynchronous dominant actor commentator model, and in a strategy network layer, carrying out the wavelet denoising, feature extraction and fusion; a robot trajectory planning scheme is optimized and calculated through an improved near-end strategy, a production line speed control scheme is obtained through a layered dual-state action value function in combination with an adaptive Coubeck-Leibler divergence constraint, an energy consumption distribution scheme is obtained based on pseudo-parallel Monte Carlo tree search and an improved soft actor commentator algorithm, and an energy consumption control scheme is obtained. The value network layer evaluates an energy consumption distribution scheme and generates a combined control strategy, after the strategy is executed, a cycle converter model is improved to evaluate production indexes, model parameters are dynamically optimized, and multi-target collaborative optimization is achieved.

The invention discloses a retrieval method and system for efficiently collecting global accurate potential customer information, and the method comprises the steps: collecting and preprocessing multi-source heterogeneous data from an e-commerce platform, social media, an enterprise official website, exhibition information, an industry database and customs; generating standardized data through spatio-temporal feature extraction and multi-dimensional credibility evaluation; establishing a dynamic topological structure by adopting a characteristic decomposition reconstruction and value network construction method; based on adaptive classification and multi-dimensional trend analysis, customer value prediction and accurate recommendation are realized; and through system quality evaluation and feedback optimization, dynamic updating and optimized display of the result are realized. According to the method, the accuracy and timeliness of customer information retrieval can be effectively improved, the information redundancy is reduced, and the reliability of the recommendation effect is improved.

The invention discloses a sparse reward environment-oriented collaborative knowledge fusion reinforcement learning method, and relates to the field of collaborative knowledge fusion reinforcement learning methods. By constructing a lightweight collaborative knowledge fusion model and a dynamic reward remodeling mechanism, the problems of low intelligent agent exploration efficiency and difficulty in strategy convergence in a sparse reward environment are solved. The method comprises the following steps: constructing a reinforcement learning framework comprising a policy network and a value network; designing an action space mutation supervision mechanism and a lightweight collaborative knowledge fusion model, and generating a smooth substitution action when a strategy is detected to be unstable; and a reward function is designed in combination with the task target and the dynamic constraint, and reward remodeling is realized by activating rewards through sub-target potential energy difference and knowledge fusion. According to the method, effective intermediate feedback can be provided for the agents in the sparse reward environment, the exploration efficiency is improved, the convergence time is shortened, the stability and cross-scene migration ability of the strategy are enhanced, and an effective solution is provided for sparse reward scenes such as robot control and multi-agent game.

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 discloses a multi-agent joint layout model training method and device, equipment and a medium, and relates to the field of printed circuit board design. The method comprises the steps that parameter information related to components to be arranged and grouping information generated based on the types of the components and aiming at the components to be arranged are acquired, and a corresponding intelligent agent is configured for each group of components; according to the parameter information and by adopting a distributed mode, carrying out layout on respective corresponding components through each intelligent agent; obtaining decision data generated by the policy network in the layout process of each agent, and evaluating the decision data by using the value network to obtain learning experience data for each component; and according to the learning experience data, determining update parameters of the strategy network and the value network of each agent by using the objective function, updating the corresponding agent by using the update parameters, and obtaining a multi-agent joint layout model based on all the updated agents. And the layout capability of the printed circuit board is improved.

The invention discloses a network information age optimization method and system based on meta-deep reinforcement learning. The method comprises the following steps: establishing a video transmission model, and collecting current state data, initial reward information and action information; the method comprises the following steps: acquiring meta-features through a meta-feature extraction function, acquiring a next-step action by an Actor strategy network in combination with a current strategy and the meta-features, sending the next-step action back to a video transmission model to give next-step state data and reward information, evaluating a next-step action effect selected by the Actor strategy network through a Critic value network, and calculating to obtain an Actor strategy network strategy gradient. Meanwhile, updating a value function of the Critic value network, setting a shearing probability ratio mechanism, and performing strategy gradient shearing on the network; and dynamically updating network parameters in combination with a meta-learning algorithm. According to the invention, the overall utilization rate of network resources is improved, and data transmission failure and retransmission caused by network congestion are reduced.

The invention discloses a power distribution network fault recovery method and system based on reinforcement learning, and the method comprises the steps: firstly carrying out the modeling of a power distribution network as a weighted directed graph model, and designing a data transmission protocol and a result receiving analysis mode of a quantum-classical hybrid architecture on this basis; meanwhile, designing a quantum reinforcement learning algorithm and strategy optimization based on a weighted directed graph model, training by using training data, representing states and actions by using quantum states, constructing and updating a strategy network and a value network through quantum gate operation, and designing a quantum reward function; when a power distribution network has a fault, the classic calculation part obtains state data after the fault according to a weighted directed graph model, encodes the state data and transmits the state data to the quantum calculation part, the quantum calculation part calculates an optimal fault recovery strategy through an algorithm and feeds the optimal fault recovery strategy back to the classic calculation part, and the classic calculation part converts the optimal fault recovery strategy into an actual control instruction to execute fault recovery work. According to the invention, efficient and reliable recovery of the power distribution network fault can be realized.

The embodiment of the invention provides a ship automatic berthing control method and system, electronic equipment and a storage medium, and belongs to the technical field of ship control. According to the method, motion state data of a current ship and a first relative distance between the current ship and a target berth are obtained, the motion state data and the first relative distance are input into a control strategy model for automatic berthing of the ship, and the ship control quantity at the next moment is obtained. The control strategy model is obtained through training based on a reinforcement learning algorithm, that is, environment interaction data is obtained, a target function is maximized, a value network is trained according to the environment interaction data, a strategy network is trained according to the value network and the environment interaction data, and the target function represents an accumulated reward value of an action output by the strategy network; the motion represents the ship control quantity, and then a control strategy model is determined according to the trained strategy network. According to the method, the workload of ship automatic berthing and leaving algorithm modeling can be reduced, the modeling precision is improved, and the safety of the ship automatic berthing process is improved.

The invention discloses a quadruped robot robust motion control method based on deep reinforcement learning, and belongs to the technical field of robot motion control, and the method comprises the steps: constructing a deep reinforcement learning model which comprises a state estimation network, a strategy network and a value network; interaction between the quadruped robot and the simulation environment is carried out, and standard observation information, historical observation information and privileged observation information of the quadruped robot at all moments are obtained; inputting the standard observation information, the historical observation information and the privilege observation information of the moment into a deep reinforcement learning model, and training based on a total loss function until convergence is carried out to obtain a trained deep reinforcement learning model; inputting standard observation information and historical observation information at corresponding moments in an actual scene into the trained deep reinforcement learning model to obtain output features of a corresponding strategy network; and the target position of each joint motor is calculated to complete the motion control of the quadruped robot. And efficient training and robust motion on various complex and unstructured terrains can be realized.

The invention belongs to the technical field of artificial intelligence and data science, and discloses a causal discovery method and related device based on a large model and structural constraint, comprising the following steps: according to meta-information of each causal variable, adopting a large language model and combining multiple rounds of dialogues, multi-role play and auxiliary cue word modes, generating an initial potential causal graph of each causal variable; repeating the following steps until a preset condition is met, and outputting a candidate causal graph with the highest value score: generating the candidate causal graph by adopting a policy network based on an encoder-decoder architecture, and carrying out iterative optimization on the candidate causal graph by adopting a reinforcement learning algorithm based on an act-critic framework; and calculating a value score of the candidate causal graph after iterative optimization according to the value network, and updating parameters of the policy network based on the encoder-decoder architecture according to the value score. The efficiency and accuracy of causal relationship recognition are remarkably improved, and a causal relationship graph with high confidence can be generated in a complex data environment.

The invention discloses a snake-shaped robot control method based on reinforcement learning. The snake-shaped robot control method comprises the steps that a complex terrain simulation scene comprising walls, cubic obstacles and low channels is built; modeling is conducted on the structure, joints and kinetic parameters of the snake-shaped robot, and a snake-shaped robot simulation model is obtained; defining a state space and an action space of the snakelike robot, and designing a diversified reward function; constructing an action network and a value network by adopting a multi-layer perceptron (MLP) structure; and performing multi-round simulation training on the snakelike robot by applying a reinforcement learning algorithm, so that the robot learns a parameter set for making an optimization decision under different scenes, thereby realizing optimization control of the snakelike robot and reaching a target point. According to the method, visual information does not need to be depended on, efficient and stable motion control of the snake-shaped robot in a complex environment can be achieved through autonomous learning, and the crossing capacity of the snake-shaped robot in a low and narrow space is improved.

Disclosed is a method for enhancement of a low-light image based on reinforcement learning and aesthetic evaluation. The method include: generating images of non-normal luminance under different lighting scenes, and constructing a training dataset for a reinforcement learning system based on the images; initializing the training dataset, a policy network, and a value network; updating, based on a no-reference reward score and an aesthetic assessment reward score, the policy network and the value network; completing model training and outputting an enhanced image result. By expanding the action space range defined in reinforcement learning, the enhancement operations for the input low-light image gain a greater dynamic range, offering higher flexibility for real-world scenarios and better meeting low-light image enhancement needs. Additionally, by incorporating the aesthetic quality assessment scores as part of the loss function, the enhanced image achieves better visual effects and higher subjective user evaluation scores.

The invention provides a motor defect identification method fusing time sequence space feature extraction and reinforcement learning. The method comprises the following steps: building a Transform-GAT model architecture T-GAT, and learning time sequence relevance and spatial topological structure features by the T-GAT to form a space-time composite representation vector; designing a dual-network architecture reinforcement learning weighted fusion mechanism of a strategy network and a value network, and after the strategy network and the value network of a parallel structure receive the composite vector, constructing a strategy function to select a defect type with the maximum probability; designing potential energy function quantization parameters, and constructing a reward function to generate a reward in combination with a difference value; taking a reward function as a target, training and optimizing model parameters of the dual-network architecture in an off-line manner, running a real-time decision in an on-line manner, and storing a tetrad to an experience pool to form a'perception-decision-feedback-update 'closed-loop mechanism; according to the method, motor defect identification is realized, downtime is reduced, and motor operation reliability and equipment operation efficiency are improved.

The invention discloses an electricity larceny risk threshold dynamic optimization and multistage early warning method based on reinforcement learning, and the method achieves the dynamic adjustment of an electricity larceny risk threshold through constructing a multi-dimensional state space containing a risk perception percentage, a line loss rate deviation, a cost budget and a load feature vector, and employing a reinforcement learning algorithm. Safety constraints are introduced to ensure the rationality of threshold adjustment, a TD3 algorithm is adopted to train and update a strategy network and a value network, the learning efficiency is improved through priority experience playback, a multi-target weighted reward function is designed to balance the multi-target optimization requirements of detection precision, line loss control, inspection cost and the like, and a red, orange and yellow three-color early warning mechanism is constructed. Early warning actions of different levels are triggered according to the ratio of the dynamically updated reference threshold value to the risk perception percentage, and early warning visualization and strategy effect evaluation are achieved through a geographic information graph dynamic rendering module, a natural language report generation module and an anti-fact analysis module.

The invention discloses a water body cleaning equipment path optimization method based on deep reinforcement learning, and the method comprises the following steps: S1, collecting water body environment data through a sensor, and carrying out the preprocessing; s2, establishing a state space and an action space for path optimization, and constructing a reward mechanism; s3, carrying out feature extraction on the preprocessed data by adopting a Transform network; s4, training the strategy network and the value network by using an improved trust region strategy optimization algorithm, optimizing an update step length of the strategy network based on KL divergence constraint, and optimizing a path planning strategy based on a strategy gradient method; s5, performing path planning by using the trained strategy network and value network; and S6, calculating the energy consumption of the water body cleaning equipment, and adjusting the operation mode or operation path of the water body cleaning equipment. According to the method, the path planning of the water body cleaning equipment is optimized by combining Transform and an improved trust region strategy optimization algorithm, and the method has the advantages of being high in environmental adaptability, low in energy consumption, high in cleaning efficiency and high in operation stability.

The invention discloses an information retrieval system and method based on semantic normalization, and relates to the technical field of artificial intelligence information, and the method comprises the steps: collecting a semantic query record input by a user, carrying out the preliminary semantic analysis, and generating structured data; on the basis of the structured data, entity disambiguation is carried out by utilizing a knowledge graph, abstract classes are generated through a neural network, calibration and dynamic weight adjustment are carried out, and high-confidence entity abstract classes and confidence scores are generated; entity abstract classes and confidence scores are combined with user contexts, an action-value function is calculated through a value network, and an optimal action is selected by utilizing a-greedy algorithm; executing semantic normalization mapping according to the optimal action, and obtaining an intermediate expression by using a meta-symbol dynamic generator; and performing index retrieval and multi-dimensional sorting based on the intermediate expression to generate a sorted retrieval result list. According to the method, the semantic fragmentation problem of multi-modal query is solved, and deep semantic alignment and dynamic weight calibration of heterogeneous data are realized.

The invention discloses a milling path automatic generation method based on reinforcement learning, and relates to the technical field of machine tool machining, and the milling path automatic generation method comprises the following steps: step 1, defining action output; step 2, defining state input including a cutter state, a global observation space and a local observation space; 3, defining a reward function; value feedback given to the intelligent agent by the environment according to the current state and action is determined through the reward function; 4, setting deep neural networks including a feature extractor, a strategy network and a value network; 5, constructing a process autonomous generation model and a machining process world model; step 6, strategy training; step 7, outputting a strategy, and autonomously completing a machining process flow; by means of the method, the problems that in the prior art, manual experience dependence is too heavy, local optimization limitation is remarkable, and physical simulation efficiency is insufficient are solved.

The invention provides a dialogue model training method, a dialogue method, a dialogue system, equipment and a medium, and relates to the technical field of man-machine interaction, the method comprises the following steps: generating current sample dialogue state information according to current sample user input information and historical sample dialogue information in a sample dialogue; performing tree search based on the to-be-trained strategy network and the to-be-trained value network to obtain current sample reply action information and a global reward value of a dialogue track corresponding to the current sample reply action information; according to the next sample dialogue state information, the current sample dialogue state information, the current sample reply action information and the global reward value which are generated by simulating the current sample reply action information, constructing sample data; and performing joint training on the to-be-trained strategy network and the to-be-trained value network according to the sample data to construct a dialogue model. According to the method, long-term return and strategy optimization are comprehensively considered during dialogues, so that the generation quality, the interaction capability and the user experience of multiple rounds of dialogues are improved.

The invention discloses a hypersonic flight vehicle adaptive attitude control method based on near-end strategy optimization, and belongs to the technical field of hypersonic flight vehicle control, and the method comprises the steps: building an air-breathing hypersonic flight vehicle longitudinal dynamic model considering power; performing reinforcement learning training on the obtained environment model by using a PPO algorithm to obtain an optimized control strategy; selecting a proper state representation and action space, and designing a strategy network and a value network for the processed model, so that an original complex attitude control problem is converted into a reinforcement learning problem which can be effectively learned; and training the reinforcement learning model by using an improved PPO algorithm to obtain a final control instruction. According to the method, the model is trained by using the PPO algorithm, so that the defects of complex controller design, strict requirement on model precision and the like of a traditional control method can be effectively overcome. A simulation experiment verifies that pitch angle tracking can be effectively realized, and the stability of an execution mechanism is guaranteed.

The invention discloses a server fault diagnosis method and device, a storage medium and a program product, and relates to the technical field of server fault localization, and the method comprises the steps: carrying out the preprocessing of original log data, determining a state space according to the obtained preprocessed log data, a fault classification model containing a value network and an action network is obtained through reinforcement learning in advance, the value network in the fault classification model is used for carrying out value evaluation on the state space, and the action network in the fault classification model is used for carrying out action probability distribution calculation on the state space according to a target value obtained through value evaluation; and determining the target fault type according to the calculated probability distribution result, thereby solving the technical problem that new features or anomalies cannot be effectively judged, effectively improving the quality and integrity of log data, improving the server fault diagnosis efficiency and accuracy, realizing intelligent identification of the fault type, and improving the fault diagnosis efficiency and accuracy. The method has the technical effects of good training stability and generalization ability.

The invention discloses a heterogeneous agent collaborative strategy training method based on a large-scale distributed hybrid unmanned and manned land battle confrontation scene. The method comprises the following steps: step a, constructing a land battle confrontation scene, a red battle agent and a blue battle agent through a VBS simulation platform; b, respectively setting a state space, an action space and a reward function of the red-party combat intelligent agent and the blue-party combat intelligent agent; c, constructing a strategy network and a value network, and inserting a cross-agent attention mechanism into the strategy network and / or the value network; and step d, through large-scale distributed training, parallel operation of a multi-instance simulation environment, strategy network fusion cross-agent information output action, value network evaluation of joint action value, training to enable the model to converge, and through a reinforcement learning algorithm, outputting an action sequence or a simulation effect to analyze a cooperation strategy of a red-party combat agent. And the cooperative combat capability of the heterogeneous agent system in a land battle confrontation scene is improved.

The invention discloses a security reinforcement learning-driven large language model security decision agent, and the decision agent comprises a high-level semantic planner which is used for receiving a target and constraint instruction in a text form, receiving a language or visual observation signal of an environment at the same time, and outputting text formatted security risk information and suggested action planning; the low-layer action actuator is used for receiving low-dimensional observation and semantic codes of the environment, and the semantic codes are output by the high-layer semantic planner after text embedding conversion; the strategy network of the low-layer action actuator outputs a final safety action; the training alignment module is used for optimizing the strategy network and the value network; a high-level semantic planner is fed back and prompted through reward and cost signals collected through environment interaction, and parameters of a strategy network and a value network are trained through a security reinforcement learning algorithm. According to the method, the decision cannot violate the given text security constraint while the decision of the given text target is completed.

The invention relates to a multi-unmanned aerial vehicle cooperative obstacle avoidance control method, device and equipment based on a PPO algorithm, and the method comprises the steps: constructing intelligent agents, deploying one intelligent agent on each unmanned aerial vehicle, and enabling the intelligent agents to comprise a strategy network and a value network; performing multi-round training on the unmanned aerial vehicles in the formation by adopting a chain-type PPO training framework to obtain path planning of each unmanned aerial vehicle and an optimal strategy of an obstacle avoidance task; and each unmanned aerial vehicle in the formation executes respective path planning and the optimal strategy of the obstacle avoidance task, so that multi-unmanned aerial vehicle cooperative obstacle avoidance control is realized. In the training process, except the unmanned aerial vehicle in the current training process, other unmanned aerial vehicles keep a fixed strategy, and the stability of the training environment is effectively kept; a chain training framework is adopted, the action amount is estimated by utilizing the local observation states of the unmanned aerial vehicles, under the guidance of heuristic information and a reward function, certain formation stability can be kept on the premise of obstacle avoidance, and the target point can be reached, so that the effect of quickly reaching the destination is improved.

The invention belongs to but not limited to the technical field of communication, and particularly relates to an air-to-ground network optimization method and system based on MEC and digital twinning, and the method comprises the steps: carrying out the learning and decision-making of an intelligent agent through employing a main strategy network and a commentator network of deep reinforcement learning; initializing parameters of a strategy network and an evaluation network, and setting training related hyper-parameters; the intelligent agent interacts with the environment according to the current strategy, executes an action and updates a state; an experience playback buffer area is used for storing experience, and updating is carried out when the capacity is reached so as to keep the latest learning experience; performing parameter updating on the main strategy network and the commentator network by adopting loss function calculation and a gradient descent method; soft updating of strategy network and value network parameters in the target network is carried out regularly to ensure learning stability; training is continued until the strategy converges, and then task allocation and trajectory planning are completed by applying the strategy.

The invention discloses a crop nitrogen fertilizer management system and method based on multi-agent reinforcement learning, and belongs to the technical field of intelligent agriculture. Comprising the following steps: collecting and preprocessing multi-source data of a target area, calibrating a crop growth-nitrogen cycle model based on the multi-source data, and constructing a dynamic simulation environment; constructing a nitrogen fertilizer application strategy model and a multi-target award function, and performing agent reinforcement learning training by adopting a centralized training-decentralized execution architecture; introducing a large language model, and updating a strategy network and / or a value network of each agent in a training process according to a reward adjustment signal and a decision constraint; when the index fluctuation ratio in the continuous evaluation period is smaller than a preset threshold value, it is judged that the nitrogen fertilizer application strategy model is converged, and an optimal nitrogen fertilizer application strategy model is obtained; and generating a nitrogen fertilizer application scheme based on the optimal nitrogen fertilizer application strategy model in combination with the real-time state data, and generating a natural language interpretation and risk assessment report based on a large language model.

The invention belongs to the field of multi-agent reinforcement learning, and discloses a relation-driven multi-agent reinforcement learning method and system based on a mixed game, and the method comprises the steps: enabling a strategy network to generate an agent action, and carrying out the interaction of an environment, so as to collect sample data; the centralized value evaluator calculates marginal influence values of the agents based on samples, and deduces a social influence weight set; assigning group external rewards as individual external social rewards based on the weights; processing the global state by using a random network distillation-driven method, updating a prediction network parameter to minimize a prediction error, and outputting an internal reward set; the external social rewards and the internal rewards are fused to form comprehensive rewards; and updating the strategy network and the value network by using the comprehensive reward, and circularly training until convergence. By adopting the method, accurate modeling of the individual interaction relationship is enhanced, invalid exploration is remarkably reduced, the learning efficiency and the strategy reliability are improved, and the overall performance of the system is enhanced through an explicit social mechanism.

The invention is applicable to the technical field of robot motion control, and provides a robot skill learning method, which comprises the following steps of: generating an auxiliary reward component function by utilizing a large language model according to learning prompt information, constructing a multi-branch value network according to a reward component function set formed by a plurality of reward component functions and auxiliary reward component functions in the learning prompt information, and performing iterative strategy training on the strategy network by using the multi-branch value network based on the weight coefficient of each value branch until the target function of the strategy network is converged, so as to obtain the target function of the strategy network. A strategy network is used for guiding robot skill learning, so that the intelligent decision-making capability of a large language model and fine-grained feedback of a multi-branch value network are combined, the strategy learning efficiency and the overall performance in reinforcement learning are improved, it is ensured that different skill obtaining requirements can be met in the strategy optimization process, and the robot skill learning efficiency is improved. And different environments and complex robot tasks can be handled, and the accuracy, precision and efficiency of task execution are improved.

The invention provides a method and system for optimizing deep reinforcement learning based on composite priority experience playback. The method comprises the following steps: acquiring a current training network and a target network; the current training network comprises a current strategy network and a current value network, and the target network comprises a target strategy network and a target value network; acquiring an experience tuple based on an action selection strategy of the current strategy network; calculating a composite priority of the experience tuple, and enabling a life cycle pointer to update an experience playback area according to the experience tuple and the composite priority of the experience tuple; and sampling from the experience playback area to update the network parameters of the current training network, and updating the network parameters of the current training network to a target network. According to the method, the time sequence difference error of the instant reward and the experience is introduced as the priority evaluation standard of the new experience, the priority of the new experience is improved through the playback times, and the replacement efficiency of the experience playback area is improved through the lifetime pointer.

The invention provides a data center IT load and cooling system cooperative control method based on a TD3 algorithm, and relates to the field of deep reinforcement learning, and the method comprises the steps: obtaining an initial data set representing the server state, environment and task characteristics of a data center, and processing the initial data set to obtain a target data set; building a deep reinforcement learning model which is based on a TD3 algorithm and comprises a strategy network Actor and a value network Critic, and determining a state space, an action space and a reward function of the model; training the model to obtain a target model; and outputting a work task allocation strategy and a cooling system regulation and control strategy through interaction of the target model and the environment. According to the target model, by outputting the work task allocation strategy and the cooling system regulation strategy, work task allocation can be guided, operation of the cooling system can be controlled, the IT load of the server and refrigeration balance of the cooling system are achieved, safe and stable operation of the server is ensured, and energy consumption of the cooling system is reduced.

The invention relates to a task unloading prediction method and device for a power grid processing task, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring an initial strategy network model and an initial value network model, and acquiring sample power grid processing task information and current training environment information; in cooperation with other edge servers and a central cloud server, according to the current training environment information and the sample power grid processing task information, performing centralized training on the initial strategy network model and the initial value network model in the edge servers to obtain a trained target strategy network model; under the condition that a new power grid processing task is received, obtaining current state information of an edge server; and inputting the current state information and the task information of the new power grid processing task into the target strategy network model to obtain task unloading prediction information for the new power grid processing task. By adopting the method, the resource utilization rate can be improved.