341 results about "Adaptive environment" patented technology

The invention discloses a multi-agent dynamic task allocation and collaborative path-finding system based on label-free distributed deep reinforcement learning. The system comprises the following steps: step 1, receiving state information and environment perception data of each agent in a multi-agent system based on distributed deep reinforcement learning; step 2, extracting feature representations of the environmental perception data and the intelligent agent state information, and performing multi-source heterogeneous information fusion through an attention mechanism to obtain state-task matching features; 3, transmitting the state-task matching characteristics to a multi-agent network in real time, realizing task allocation negotiation among agents by adopting a hierarchical scheduling and state exchange mechanism based on task priorities, dynamically detecting newly added task types, and performing incremental learning; the online updating iteration of the model is realized to assist the multi-agent optimization task allocation strategy and the path planning action; compared with the prior art, the method has the advantages that by applying the distributed deep reinforcement learning technology and a state exchange mechanism between intelligent agents, the system can quickly adapt to environment changes and task dynamics, and the resource utilization rate and the task completion efficiency are improved.

The invention discloses a servo motor dynamic overload detection and protection method based on a heat accumulation model, which relates to the technical field of servo system fault protection application, and comprises the steps of constructing a heat accumulation equivalent model, dynamically detecting an overload state, executing a heat dissipation recovery strategy, implementing a multi-stage protection mechanism, self-adaptive environment compensation and a historical data learning function. Through discretization processing of overload curve data and a nonlinear acceleration calculation strategy, the problems of protection lag and false triggering caused by traditional single-parameter detection are avoided, it is ensured that the motor can respond in time in light load, heavy load and load fluctuation scenes, the burn-out risk of the motor is reduced, and through integration of an environment temperature sensor and a heat dissipation condition evaluation unit, the reliability of the motor is improved. The heat accumulation coefficient and the heat dissipation attenuation parameter are dynamically adjusted, the problem of model deviation caused by environment interference is solved, when heat dissipation deterioration and temperature feedback abnormity are detected, the heat dissipation recovery period is automatically prolonged, calculation logic is corrected, and the continuous operation reliability of the system is improved.

The invention relates to artificial intelligence generation content AIGC, in particular to an AIGC test case self-adaptive generation system based on risk feedback, which comprises a semantic variation generation module for selecting a variation operator from a variation operator library according to a self-adaptive strategy, performing variation processing on a seed case provided by a basic test case library by using the variation operator, and generating a semantic variation result; generating a new test case; the AIGC model takes the test case as model input and performs model output to the risk assessment and quantification module through an AIGC model interface; the risk assessment and quantification module is used for performing multi-dimensional risk assessment on a model output result by utilizing the risk detection model library, calculating a risk level and generating a test report; the risk feedback processing module is used for storing the high-risk test case, performing feature extraction and converting a risk assessment result into an operable risk feedback signal according to extracted features; according to the technical scheme provided by the invention, the defects of low efficiency and difficulty in automatically adapting to environment change can be overcome.

The invention discloses an intelligent abnormity early warning system for an industrial Internet of Things edge node, which comprises an edge sensing layer, an edge intelligent layer and a cloud edge collaboration layer, and is characterized in that the edge sensing layer is used for data acquisition and security preprocessing; the edge intelligent layer is used for intelligently analyzing the data transmitted by the edge sensing layer and constructing a dynamic space-time diagram neural network model and a dynamic edge knowledge graph; a digital twinning threshold engine is set, when the digital twinning threshold engine detects abnormity, reverse reasoning is carried out along the knowledge graph, and self-adaptive early warning is carried out; and the cloud edge collaboration layer is used for carrying out collaboration processing on early warning triggered by the digital twin threshold engine. According to the invention, the spatial-temporal features of the equipment group are learned through the set dynamic spatial-temporal diagram neural network; the digital twin threshold engine generates a dynamic threshold based on physical simulation and adapts to environment and load changes; the edge layer realizes lightweight real-time analysis; and a full-link embedded chaotic encryption and national cipher SM4 algorithm is adopted.

Techniques for adjusting vehicle models based on environmental conditions are discussed herein. The techniques may include receiving image data representing a portion of an environment in which a vehicle is operating and inputting the image data into a machine learned model. Additionally, data representing an environmental condition associated with the environment may be received from a sensor of the vehicle to detect changes in the environmental conditions such that one or more actions associated with the machine learned model or an output of the machine learned model may be performed. Some of the techniques may also include running multiple machine learned models or multiple configurations of a machine learned model in parallel and selecting different outputs of the machine learned model(s) to adjust for changes in the environmental conditions. For instance, individual outputs may be selected based on environmental conditions, confidence scores, thresholds, etc.

The invention provides an RFID label positioning error correction method based on deep learning, and the method comprises the following steps: 1, deploying a multi-mode sensing network in a target region, synchronously collecting radio wave signals and environment auxiliary data of an RFID label, and constructing a space-time multi-dimensional feature data set; step 2, constructing a joint architecture hybrid model composed of a space-time Transform network and a generative adversarial network; according to the method, a space-time multi-dimensional feature data set is constructed by fusing RFID signals and environment auxiliary data through a multi-mode sensing network, adaptive parameters are dynamically generated in combination with a model-independent meta-learning algorithm so as to quickly respond to environment changes, multi-path signal correlation is captured by using a space-time Transform and generative adversarial network combined architecture, feature expression is optimized, and the robustness of the system is improved. The problems that a single signal feature is missing, model parameters cannot dynamically adapt to the environment and the anti-interference robustness of a traditional network is insufficient are effectively solved, and the effectiveness, the real-time performance and the precision of RFID label positioning in the complex environment are remarkably improved.

The invention relates to the technical field of equipment monitoring, in particular to an anomaly monitoring method for water quality detection equipment, which realizes multi-level anomaly detection by deploying a plurality of monitoring nodes in a target water area, constructing a distributed monitoring topological structure and combining an auto-encoder, an isolated forest and an LOF algorithm. Based on historical data, a reference model of normal equipment is established, Apache Kafka is used for collecting data streams in real time, and standardization and windowing processing is carried out through Flink. A reconstruction error is calculated based on an auto-encoder, a feature matrix is constructed, global abnormal points are screened by using an isolated forest, local outlier is verified through an LOF algorithm, and an abnormal state is comprehensively judged. The method supports incremental learning, adapts to environmental changes, has the characteristics of high real-time performance and low delay, can effectively identify abnormal conditions such as sensor aging and measurement errors, improves data reliability and system stability, and remarkably reduces maintenance cost.

The invention discloses an internal and external network security service passing method and system based on edge computing, and belongs to the technical field of network security, and the method comprises the steps: obtaining multi-dimensional attribute information of an edge node, generating a security policy basic data set, generating a traffic feature fingerprint and a self-adaptive environment adaptive fingerprint, and calculating a fusion matching degree; edge security entity nodes are generated in combination with service scene adaptive threshold clustering, and then a security policy meta-model with a security policy blueprint as a core is constructed; analyzing the security policy meta-model through a multi-modal semantic analysis engine, generating a dynamic security enhancement model, and mapping the dynamic security enhancement model to a hierarchical security control model; based on the hierarchical model, outputting edge node internal and external network safety passage configuration and a corresponding safety passage ledger through a scenarized configuration generation algorithm; according to the method, adaptive generation, dynamic optimization and accurate execution of the security policy are realized, and the security, the automation level and the operation and maintenance efficiency of internal and external network passing in the edge computing environment are effectively improved.

The invention relates to an intelligent power grid cooperative scheduling method for automobile access, and relates to the field of electric automobile charging scheduling and intelligent power grid optimization. The method comprises the following steps: acquiring multi-source real-time data of an electric vehicle, a power grid and a charging station, constructing a collaborative scheduling graph structure, and performing multi-target optimization to generate an initial charging guide strategy and a charging station resource allocation scheme; then, reinforcement learning iteratively optimizes the guide strategy to dynamically adapt to environmental changes; further optimizing charging time and power parameters through a dynamic adaptive optimization algorithm, and realizing charging station congestion early warning and selection suggestions in combination with a probability prediction algorithm; a charging station power distribution strategy is optimized based on early warning information, fine optimization is carried out by adopting a genetic algorithm, and a scheduling scheme is evaluated and adjusted in real time through a feedback control algorithm. According to the invention, dynamic, refined and global optimization of electric vehicle charging scheduling is realized, the stability of a power grid, the operation efficiency of a charging station and the charging experience of a user are improved, and new energy consumption is promoted.

The invention discloses a multi-scene wireless network rapid deployment method based on an electromagnetic field digital twinning technology, and relates to the technical field of computer application, and the method comprises the following steps: S1, constructing a three-dimensional digital twinning model of a corresponding scene, and integrating a building structure, equipment layout and material attribute data, and establishing a scene special material database containing metal equipment, a concrete wall body and an electromagnetic shielding layer. According to the multi-scene wireless network rapid deployment method provided by the invention, the three-dimensional digital twin model is constructed, and the building structure, the equipment layout and the material attribute data are integrated, so that the precise simulation of the environment is realized, the wireless network deployment accuracy is improved, the deployment scheme can better adapt to the dynamic change of the environment, and the deployment efficiency is improved. The improved O-FDTD algorithm is adopted to simulate electromagnetic field propagation in a building in real time, and the problem that the calculation efficiency is low when a traditional FDTD algorithm processes the metal edge effect and a moving object is effectively solved.

The invention relates to the technical field of hydrological data processing, in particular to a water conservancy intelligent internet-of-things sensing method and system, and the method comprises the following steps: combining statistical data into a feature set based on hydrological monitoring data including flow, water level and rainfall information, judging whether the data accords with normal distribution, and generating a data distribution feature vector set. According to the method, the change trend of the data is captured through multi-level data distribution characteristic analysis, and the monitoring precision of the data exception is enhanced. In the prediction process, new data and weight adjustment are gradually fused, so that the model can adapt to environmental changes in real time, and the hydrological state in a future short period is accurately predicted. Besides, abnormal data can be identified in real time based on analysis of quantiles, potential problems can be quickly found, and state division of different levels can be carried out according to abnormal degrees. In sampling scheduling, the priority of data acquisition is adjusted according to a probability model of state transition, so that monitoring resources are efficiently configured.

The invention provides a multi-agent-based task cooperative execution method and device, and the method comprises the steps: carrying out the calculation based on the task type of a target task and the current system state of a multi-agent system, and obtaining an index weight; determining a task allocation scheme of each subtask based on the index weight and the agent bidding information of each subtask; and determining an execution strategy of each sub-task based on the task attribute of each sub-task, the task allocation scheme and the environment perception data of the multi-agent system. According to the method and device provided by the invention, the index weight is calculated according to the task type of the target task and / or the current system state of the multi-agent system; the task allocation scheme of each subtask is determined based on the index weight and the bidding information of each subtask corresponding to the plurality of agents, so that dynamic task allocation adapting to environment change is realized, the task allocation rationality is improved, and the task execution efficiency and task completion quality of the system are improved.

The invention relates to the technical field of tourism big data, in particular to a multi-modal tourism route personalized generation method and system based on deep learning, and the method comprises the steps: constructing and incrementally updating a tourism knowledge hypergraph, extracting hidden time sequence preference and cross-entity cooperation signals of a user through a space-time perception graph neural network, and generating an interest drift model; performing deep fusion and intention analysis on the interest drift model and a context signal sensed in real time, and decoding to generate a candidate route concept skeleton by introducing an attention competition mechanism; and carrying out multi-dimensional simulation deduction and adaptability evaluation on the candidate route concept skeleton, carrying out iterative optimization through a reinforcement learning strategy based on deduction feedback, and finally outputting a personalized tourist route with optimal robustness, so that the robust tourist route which is highly personalized and adapts to environmental changes can be generated, and the robustness of the tourist route is improved. Therefore, recommendation accuracy and user experience are improved.

The invention relates to the technical field of environment perception, and discloses a fusion perception system and method based on a multispectral sensor. According to the method, asynchronous heterogeneous environment data streams collected by a visible light sensor, a thermal imaging sensor, a millimeter-wave radar and a laser radar are obtained, environment information of multispectral bands is covered, comprehensive capture of target appearance, temperature, motion and geometric characteristics is achieved, space-time registration operation is executed on asynchronous heterogeneous data, space-time deviation between the sensors is eliminated, and the accuracy of target tracking is improved. And a multi-source sensing data cube with time-space synchronization is generated, and a consistent data basis is provided for subsequent fusion. Based on the data cube, a hierarchical feature fusion architecture is adopted to extract cross-spectrum joint features, a multi-dimensional feature tensor is generated, and internal association and complementary information of multi-source data is deeply mined. The multi-dimensional feature tensor is processed through a dynamic weighting decision mechanism, a three-dimensional situation awareness map of an environment target is output, environment dynamic change and sensor data fluctuation are adapted, and the reliability of environment awareness in a complex scene is improved.

The invention discloses a power transformation and distribution station room operation method and device based on a force control dexterous hand, and belongs to the technical field of robots. According to the technical scheme provided by the embodiment of the invention, the distributed multi-source sensor array is utilized to obtain the consistent environment data, the operation situation description is dynamically generated to adapt to the environment change, the complete operation scheme is intelligently generated in combination with the cross-domain strategy knowledge graph, and real-time feedback is performed through the second multi-mode sensing data flow acquired during control, so that the control accuracy is improved. The problems of low efficiency and safety risk coexistence in the operation of the power transformation and distribution station room are solved.

The invention relates to the technical field of robot control, and discloses a robot cerebellum reinforcement learning method and system based on adaptive environment change, and the method comprises the steps: collecting multi-modal sensing data in the operation process of a robot; based on the multi-modal sensing data, determining an environment sudden change increment and an environment characteristic variable quantity, and obtaining a main strategy network according to the environment sudden change increment and the environment characteristic variable quantity; outputting a basic action vector of the robot based on the main strategy network, and performing value evaluation on the state of the robot to obtain a value evaluation quantity; performing updating calculation on the basic action vector based on the value evaluation quantity to obtain a compensation action vector; and performing vector superposition fusion on the basic action vector and the compensation action vector to obtain a final control vector, and taking the final control vector as a cerebellum learning action instruction of the robot, thereby realizing dynamic adaptation of a reinforcement learning strategy and cerebellum rhythm control. And the motion stability and the learning convergence efficiency of the robot in the unstructured environment are optimized.

The invention discloses a multi-period time difference weighted ranging method based on obstacle perception, which relates to the technical field of intelligent perception and sensor fusion, and specifically comprises the steps of multi-period time difference measurement, obstacle attenuation coefficient modeling, weighted fusion calculation, dynamic error compensation and final distance output. According to the multi-period time difference weighted distance measurement method based on obstacle perception, through multi-period time difference measurement and weighted fusion, random errors are reduced, and the distance measurement precision is effectively improved; by introducing the number of obstacles as a key variable, the weight of a distance measurement formula is dynamically adjusted to adapt to a complex scene, and dynamic environment adaptation is realized; and through combination with an obstacle attenuation model, the influence of multi-path interference on a distance measurement result is effectively suppressed, and the anti-interference capability is effectively enhanced.

A building-integrated artificial intelligence system forming a continuous ambient neural field is disclosed. The system includes a distributed multimodal sensor lattice, an on-premise symbolic cognition engine, and an adaptive environmental control kernel operating entirely at the building edge without reliance on external cloud services. Sensor data from optical, thermal, acoustic, airflow, pressure, structural, electrical, and chemical modalities are transformed into non-identifying occupancy vectors, behavioral glyphs, risk indicators, and environmental state descriptors. A privacy-governed policy graph determines sensor permissions, redaction thresholds, consent conditions, emergency overrides, and jurisdiction-specific compliance parameters. The neural field predicts occupancy loads, optimizes HVAC, ventilation, and lighting, detects accidents and structural anomalies, classifies emergent risks, and generates redacted event capsules for audit and emergency dispatch. A federated topology enables multiple buildings to exchange compressed symbolic templates to improve predictive accuracy without transmitting raw data. The system provides a universal, regulation-aligned AI nervous system for autonomous building operations.

The invention provides a robot cluster deployment method and system, electronic equipment and a medium. The method comprises the steps of obtaining task complexity information and environment resource information of a to-be-allocated task; according to the attribute information of the robots in each robot alliance in the robot cluster, through a load balancing algorithm, selecting a target robot alliance for executing the to-be-allocated task in each robot alliance; according to the task complexity information, the environment resource information and the state information of each robot in the target robot alliance, adopting a comprehensive evaluation algorithm of task allocation to obtain the fitness of each robot in the target robot alliance for executing the to-be-allocated task; the to-be-allocated task is issued to the target robot with the highest fitness in the target robot alliance, the target robot is used for executing the to-be-allocated task, intelligent allocation of the task is achieved, the allocation scheme can adapt to environment changes, and the task execution efficiency and the resource utilization rate of the robot cluster are improved.

The invention relates to the technical field of photovoltaic power generation detection, in particular to an outdoor photovoltaic module attenuation rate test method and test system, and the method comprises the steps: a multi-dimensional environment and operation data collection process; constructing and updating a digital twin reference model; a self-adaptive environment compensation and normalization process; and attenuation rate calculation and state evaluation. Nonlinear measurement errors caused by spectrum red shift / blue shift, large-angle incidence and a temperature hysteresis effect are effectively stripped, and real-time output data are mapped to equivalent performance data under a standard test condition, so that the measurement uncertainty is remarkably reduced, and sensitive capture and accurate calculation of a micro power attenuation trend are realized.

The invention discloses a heterogeneous terminal behavior baseline auditing system and method based on artificial intelligence, and the method comprises the steps: carrying out the time sequence modeling and clustering analysis of large-scale and diversified traffic data collected by bypass monitoring or network equipment through introducing a streaming machine learning model supporting an online learning capability, and carrying out the clustering analysis of the data, normal behavior state mode clusters of various terminals are effectively extracted and dynamically maintained; meanwhile, in the application stage, a deviation measurement network is combined, real-time calculation of a deviation score between newly generated update behavior data and a historical normal mode is achieved, and a concept offset signal and an adjustment mechanism are automatically triggered accordingly, so that the system can continuously adapt to environmental changes. Therefore, the defects of the traditional scheme in generalization ability, self-adaptive ability and real-time performance are overcome, and the new requirement for efficient and safe management of the heterogeneous terminal in the current complex and changeable informatization environment can be better met.

The invention relates to a multi-unmanned aerial vehicle collaborative decision-making method, device and equipment based on a graph attention network. The method comprises the following steps: constructing a spatial topological graph by taking a defense unmanned aerial vehicle and an intrusion target as elements; in the spatial topological graph, determining a current agent unmanned aerial vehicle, and a neighbor defense unmanned aerial vehicle and a neighbor intrusion target adjacent to the current agent unmanned aerial vehicle; constructing a multi-unmanned aerial vehicle collaborative decision-making model, and calculating high-dimensional features of the current proxy unmanned aerial vehicle; based on the position and speed information, calculating a first aggregation feature of the neighbor defense unmanned aerial vehicle and a second aggregation feature of the neighbor intrusion target through an attention weight mode; splicing the high-dimensional feature, the first aggregated feature and the second aggregated feature to obtain a fused feature; and generating an action mean value based on the fusion features, assuming that the action mean value obeys Gaussian distribution, and sampling from the Gaussian distribution to generate the action of the current agent unmanned aerial vehicle. The collaborative decision-making efficiency can be improved, and the method can quickly adapt to environment changes.

The invention discloses a safety decision-making method for patrol scheduling of multiple unmanned vehicles in a confrontation environment. The safety decision-making method comprises the following steps: S1, acquiring environment perception information through a scene evaluation layer; s2, inputting the environmental perception information into a decision-making layer to carry out inspection scheduling planning based on reinforcement learning under a hierarchical decision-making execution mechanism; and S3, the decision-making layer monitors the decision execution quality in real time and carries out automatic recovery and retry. By adopting the technical scheme of the invention, a multi-unmanned vehicle system can better adapt to environmental changes in a large-scale task dynamic scene, the defect of robustness of a traditional multi-agent reinforcement learning algorithm is overcome, and efficient completion of preset inspection performance indexes is realized.

The invention discloses an agent intention understanding method based on multi-modal information fusion, relates to the technical field of natural language processing, and solves the technical problem that multi-agent intention understanding cannot be realized in a dynamic environment. Voice, text and visual features are mapped to the same hidden space through maximum mean value difference minimization, distribution difference between modes is eliminated, joint features in the hidden space can capture cross-mode complementary information at the same time, and the richness and anti-noise capability of feature expression are enhanced. CRF (Conditional Random Field) is combined with a special dictionary in the scheduling field to carry out candidate set expansion on fuzzy keywords, and specific semantic variants in the field are captured. The optimal grammar of the scheduling instruction is analyzed and decomposed into a feature vector causal relationship pair, and components of an intention and logic dependence of the intention are clarified. A dynamic knowledge graph is constructed based on the causal relationship, probability dependence among parameters is quantized by using a Bayesian network, the causal relationship strength is dynamically adjusted, and environmental changes are adapted.

The invention provides a sensor data processing method, a robot control method, a sensor data processing device and a computer storage medium. The sensor data processing method comprises the following steps: determining self-adaptive environment compensation parameters of original sensor data according to environment data; determining an adaptive time compensation parameter of the original sensor data according to the sensor speed; obtaining a dynamic compensation parameter of the original sensor by using the self-adaptive environment compensation parameter and the self-adaptive time compensation parameter; performing dynamic compensation on the corresponding original sensor data according to the dynamic compensation parameters of the original sensor data; and the multiple pieces of original sensor data after dynamic compensation are fused, and fused sensor data used for analyzing the information of the environment where the robot is located are obtained. Through the sensor data processing method, self-adaptive dynamic preprocessing of the sensor data by using the environment data and the motion data is realized, so that the sensor data is calibrated, and the accuracy of the sensor data is improved.

The invention provides a self-adaptive operation strategy optimization method and system for a refrigeration system, and relates to the technical field of automatic control, and the method comprises the steps: obtaining temperature and humidity data and fluid boundary parameters, reconstructing a temperature and humidity distribution field through a physical constraint neural network, extracting enthalpy difference features, constructing a topological coupling network, and calculating and correcting a refrigeration load. And on the basis of energy consumption and enthalpy value uniformity, refrigeration resource distribution parameters are optimized, accurate regulation and control of the dynamic load area are achieved, the energy efficiency of the system is improved, the enthalpy value uniformity requirement of each area is met, and the environment change adapting capacity of the system is enhanced.

The invention relates to an intelligent ship network optimization method, system and device based on multi-link convergence and a medium. The method comprises the following steps: firstly, acquiring multi-network link connection parameters of the intelligent ship, constructing an adaptive environment, modifying a protocol stack, and integrating a physical link into a virtual link to realize resource convergence; and based on the real-time parameters of the virtual link and the physical link, dynamically distributing traffic by adopting an improved aggregation protocol, and generating a load balancing scheme. And then, regularly sending a detection packet quality index, and predicting a link quality trend in combination with historical data and navigation longitude and latitude. And finally, dynamically adjusting flow distribution according to a prediction result, an application real-time demand, a bandwidth demand and a current position tariff characteristic, and generating a ship-shore communication scheme considering both efficiency and cost. By adopting the method, the stability, reliability and economy of ship-shore communication of the intelligent ship can be effectively enhanced, and the method is suitable for multi-scene network monitoring optimization.

The invention relates to the field of biological fermentation engineering and discloses an environment-adaptive fermentation process temperature adjusting method which comprises the following steps: acquiring temperature field data in a fermentation tank through a sensor array to obtain initial temperature distribution data; performing microbial activity associated thermodynamic feature extraction on the initial temperature distribution data to obtain fermentation thermodynamic feature data; according to fermentation thermodynamic characteristic data, spatial-temporal distribution density analysis of temperature gradient difference is carried out, and temperature gradient spatial-temporal distribution density data is obtained.According to the method, through a depth deterministic strategy gradient algorithm, a self-adaptive temperature adjusting model is constructed, closed-loop temperature control of the fermentation process is achieved, the model can automatically adjust the temperature according to real-time data, and the temperature control accuracy is improved. The dynamic change in the fermentation process is responded, the adjustment precision is improved, the provided self-adaptive fermentation temperature adjustment method can achieve refined and intelligent temperature adjustment and control in the fermentation process, the fermentation efficiency is improved, and the method has wide application prospects.

The invention discloses a target-driven energy supply and demand intelligent optimization system and method. The system comprises an input module, a multi-target quantification module, a decision module, an execution module and a feedback module. The input module receives an input high-level target instruction; the multi-target quantification module receives a target instruction to form a target quantification signal; the decision module is connected with the dynamic multi-target quantization module and is used for receiving the target quantization signal and outputting a strategy execution signal; the execution module is connected with the reinforcement learning intelligent decision-making module and is used for converting the strategy execution signal into a control instruction of the energy equipment; the feedback module collects energy system state data and external environment data in real time, and drives target weight adaptive adjustment and strategy iteration updating. According to the target-driven energy supply and demand intelligent optimization system and method, the problems that an existing energy optimization system is difficult to dynamically convert an abstract high-level target into a quantization parameter and cannot adapt to environmental changes to adjust the priority of the target can be solved.

The invention discloses a data center active preventive operation and maintenance method based on intelligent group control, and relates to the technical field of telecommunication technologies. Comprising the steps of collecting data center environment parameters and equipment operation parameters in real time and performing data preprocessing; preliminarily defining a state space, and constructing a state incidence matrix and a state space model between devices; defining a control action space, and establishing an optimization function by taking minimization of energy consumption efficiency and avoidance of equipment overheating as targets; a state space is improved by fusing a historical mean value and a real-time state and quantifying a system dynamic trend, an intelligent agent is trained based on Markov decision, and a dynamic adjustment strategy is embedded; the intelligent agent generates a control instruction, and the experiment platform executes the instruction and feeds back the instruction to form a dynamic closed loop. The problems that in the prior art, real-time dynamic adjustment is insufficient, and the operation and maintenance effect and the response capacity are poor are solved, and the purposes of dynamic adjustment, environment and parameter change adaptation, high fault recognition rate and high response capacity are achieved.