668 results about "Priority queue" patented technology

The invention discloses a distribution line fault intelligent diagnosis method based on multi-source information fusion, and belongs to the technical field of power system fault detection and diagnosis, and the method comprises the steps: obtaining distribution line data and meteorological data; superposing the distribution line data with the meteorological data to generate a space-time fusion feature matrix based on the wire space displacement; dynamically adjusting a preset initial threshold value by using the space-time fusion feature matrix; and performing multi-round verification on the abnormal fluctuation of the electrical parameters of the line according to the fault judgment threshold, generating a fault positioning instruction, performing real-time matching with the topological structure of the line, correcting the priority queue of the fault section based on the switching action feedback data of the feeder terminal, and triggering the fault removal action matched with the priority. According to the method, spatial-temporal feature fusion of current, voltage, weather and spatial displacement data and a dynamic threshold adjustment mechanism are adopted, the fault type can be accurately identified, the fault section can be positioned in real time, and the diagnosis precision and the response speed in a complex environment are remarkably improved.

The invention provides a computing resource optimization method and system for analysis tasks, and belongs to the technical field of computers.The computing resource optimization method comprises the steps that the priority of each analysis task is determined according to feature information of each analysis task and current available computing resource state information of the system, and a priority queue is generated; obtaining a task load prediction value according to the historical task load data and the real-time system state data; according to the type of the analysis task, the data scale and the data source position, the analysis task with the real-time requirement higher than a preset standard is allocated to an edge node to be executed, and resource allocation of the edge node is dynamically adjusted according to a task load predicted value; and dynamically allocating available resources from the computing resource pool according to the priority queue and the task load prediction value so as to execute the plurality of parallel analysis tasks. According to the resource management method based on task priority dynamic adjustment, task load prediction and edge computing optimization scheduling, real-time prediction and dynamic self-adaptive scheduling of computing resources are achieved.

The invention relates to the technical field of information retrieval, and discloses a distributed storage and indexing method, which comprises the following steps of: dynamically fragmenting an original file through an improved consistent Hash algorithm to generate a plurality of data blocks with timestamps; constructing a three-dimensional Bloom filter index matrix containing timestamps, data types and content features for the data blocks with the timestamps, associating a B + tree local index with an inverted global index through a hierarchical index structure, and establishing an index update priority queue by adopting a two-channel synchronization mechanism, performing real-time increment synchronization on the hotspot index through a heartbeat mechanism, performing batch synchronization on the cold data layer index according to a cold data synchronization period, predicting a data distribution probability through a distributed query statistics probability table during query, and initiating multi-path query in parallel based on probability weight, and the query path is dynamically optimized according to the node group storage medium type and the inter-node network transmission delay, so that rapid distribution adjustment and access of distributed storage are realized.

The invention relates to the technical field of data processing, and discloses an automobile manufacturing industrial data distributed processing method and system based on digital twinning. The method comprises the following steps: monitoring and collecting production abnormal events, decision emergency degree parameters and simulation task types based on service states to generate a service priority evaluation data set, and performing priority ranking on a plurality of simulation tasks to form a dynamic priority queue and a resource demand matrix, the queue is input into a distributed simulation engine for CPU, memory and GPU load balancing processing to generate a node resource allocation scheme, and preemptive scheduling is executed to migrate low-priority tasks to idle nodes to form a task execution mapping table; and monitoring simulation progress and resource consumption through an adaptive scheduling mechanism to obtain scheduling performance feedback data, and updating the dynamic priority queue. According to the method and the device, the problem of lack of an intelligent scheduling mechanism during concurrent execution of multiple simulation tasks in the prior art is solved, and the response speed of the key simulation task and the utilization efficiency of computing resources are improved.

The invention provides an AI-based laboratory equipment scheduling optimization method and system, and the method comprises the steps: firstly obtaining a state monitoring data set containing the characteristics of equipment operation power consumption, idle time length, environment interference factors and the like in real time, and then carrying out the multi-dimensional analysis of the state monitoring data set; generating an availability evaluation index set containing characteristics of equipment load fluctuation, maintenance period prediction, compatibility matching and the like, and an experiment task priority queue, performing cross decision analysis on the availability evaluation index set and the experiment task priority queue based on a preset dynamic resource allocation model, and obtaining a scheduling strategy set containing a task allocation path, a cooperative operation rule and a conflict resolution mechanism; scheduling strategy parameters are calibrated according to experimental task operation log data, an optimized execution instruction set is generated, instructions are fed back to an equipment control system to adjust the equipment state, a dynamic resource allocation model is iteratively updated according to execution feedback data, and efficient scheduling optimization of laboratory equipment is achieved.

The invention relates to the related technical field of digital data processing, in particular to a personalized federated learning method and system for a heterogeneous multi-source industrial internet, and the method comprises the steps: connecting a client, evaluating a load, time delay and modal similarity to generate a dynamic association table, deploying a hierarchical encryption protocol, and constructing a priority queue; a cache mechanism is set to coordinate distributed iterative optimization, so that the technical problem that network oscillation and computing resource waste are aggravated due to overhigh load of part of nodes caused by frequent access and exit of equipment and data volume difference in the industrial internet and repeated migration of clients and nodes is caused is solved, cross-equipment shared knowledge base vectors are extracted, and the computing efficiency is improved. The technical effects of reducing the influence of model isomerism on aggregation, dynamically scheduling high-frequency parameter local aggregation and low-frequency parameter cloud synchronization, optimizing the association weight of a client and a fog node in real time, realizing privacy protection and efficient personalized federated learning, and ensuring the privacy and security of user data in the training process are achieved.

The invention provides a network service acceleration method and system applied to a security transaction service platform, and the method comprises the steps: firstly obtaining a security transaction request set sent by a user terminal in a transaction period, enabling each transaction request unit to comprise a request type identifier, a timestamp and content data, and then carrying out the classification according to the request type identifier to generate a priority queue, the method comprises the following steps: determining a queue ordering rule according to a timestamp, dynamically screening a proper service node group from a pre-established distributed service node cluster, mapping and binding the proper service node group with a priority queue, distributing request units to corresponding service node groups in batches according to the ordering rule to execute transaction operation, generating a response data packet, and returning the response data packet to a user terminal. And when the service node group executes the operation, the load state parameters are updated in real time and synchronized to the global load monitoring module, so that the network service acceleration performance of the security transaction service platform is effectively improved, and efficient transaction is ensured.

The invention relates to a high-fidelity cloud rendering cluster scheduling method and system, and the method comprises the steps: receiving a rendering task, generating a multi-level priority queue according to the task complexity, timeliness and resource demand classification, and automatically optimizing a scheduling strategy; monitoring the resource state of the heterogeneous computing node in real time; allocating tasks by using preemptive and round-robin scheduling strategies, and adjusting and coping with resource fluctuation in combination with a dynamic code rate; the tasks are decomposed by adopting a spatial blocking and time framing strategy, and an execution sequence is controlled according to a topological sorting algorithm; abnormal nodes are identified through heartbeat detection, and affected subtasks are migrated through incremental task updating. According to the method, efficient resource allocation, scheduling algorithm and idle key frame scheme can be realized, the GPU directly outputs the video stream, the rendering speed is remarkably improved, resource waste and operation cost are reduced through a dynamic resource allocation mechanism, a fault-tolerant mechanism is provided, the task is ensured to be normally completed under the condition of node fault, and the task efficiency is improved. And distributed rendering and synthesis of large-scale high-resolution images are supported.

The invention relates to the technical field of cloud side end computing, and discloses a cloud side end cooperative computing resource dynamic allocation method based on task-aware migration. The method comprises the following steps: acquiring real-time load characteristics and resource demand characteristics of calculation tasks in a cloud side end system, and dividing task priority queues in combination with task type identifiers; extracting historical execution records of tasks at cloud, edges and terminal nodes, constructing a task execution feature library, and generating a resource demand prediction model in combination with real-time load features; analyzing network transmission time delay characteristics of a cloud end and edge nodes, measuring real-time calculation capability fluctuation data of terminal equipment, and establishing an inter-node resource collaboration degree evaluation matrix; generating an initial migration strategy according to the prediction model and the evaluation matrix, monitoring actual resource occupancy deviation of the task, forming a final decision in combination with a node resource state correction strategy, triggering cross-node migration, and synchronously updating the priority queue and the evaluation matrix.

The invention discloses an intelligent private cloud resource scheduling management system based on a hyper-converged architecture, particularly relates to the technical field of virtual resource scheduling management, and is used for solving the problem of overall utilization rate reduction caused by cascade resource oscillation and fragmentation due to fault recovery in the prior art. Candidate buffer nodes are screened based on the resource occupancy state, a migration priority queue is generated in combination with the node resource isomerism matching degree and the real-time load, and an independent buffer resource pool is reserved at a target node; key resources are directionally reserved through resource locking labels, task dependence periods are associated, and task conflict risks possibly triggered after the resources are released are dynamically predicted; adjusting a label effective period to cover a task execution window based on a conflict identification result; and removing the label in the preset time window, releasing the resources to the buffer pool, synchronously triggering global resource integration and candidate node list updating, and realizing efficient recovery and dynamic redistribution of fragmented resources, thereby inhibiting multi-node resource fluctuation.

The invention relates to the technical field of artificial intelligence data centers, and discloses a computing power resource scheduling method and system for an AI multi-service data center, and the method comprises the following steps: S1, obtaining multi-service task request data, processing the task request data based on a time sequence analysis model, and generating resource demand prediction data; s2, according to the resource demand prediction data, heterogeneous computing resource data is converted into virtual resource pool data, and priority queue data is generated based on task priority data; s3, processing the virtual resource pool data and the task priority data based on a reinforcement learning algorithm to generate resource allocation strategy data, and processing high-frequency access data through a prefetching strategy; and S4, generating edge-cloud hierarchical scheduling data according to the task delay demand data and the network topology data. According to the method, heterogeneous resources are dynamically allocated through time sequence analysis and a reinforcement learning algorithm, the resource utilization rate is increased to 85% from 60%, and the vacancy rate is reduced to 5%.

The invention relates to a distributed outbound system optimization method based on cloud computing. The method comprises the following steps: disassembling an outbound task into fine-grained subtasks including dialing, recording, interaction and recording; generating a unique identifier for each subtask; based on a distributed task scheduler; according to node real-time performance and network delay, dynamically distributing tasks, and introducing a task priority rule; monitoring a task execution state in real time through a distributed tracking system; recovering and redistributing abnormal and failed tasks; fragmenting the call-out data according to regions or customer groups, and distributing the call-out data to different nodes; processing the call record and the client feedback by using a stream processing framework comprising Apache Flink; adding a call task including Kafka into a distributed task queue; setting a dynamic priority queue, and adjusting the priority according to the task timeliness; setting an overtime threshold value for the call task, and monitoring a task state in real time; and when the task is overtime or failed, triggering a retry strategy including adjusting the call period and switching the node.

The invention provides a mutual inductor test data cloud edge cooperative processing method and device, and relates to the technical field of data processing, and the method comprises the steps: carrying out the time domain and frequency domain combined feature extraction of the original measurement data of a mutual inductor test through a multi-mode decoupling preprocessing model, and forming a data feature vector set, further generating a confidence label flow through state recognition and Bayesian inference, evaluating a prediction error, and realizing data classification screening and priority queue construction; the data are uploaded to a cloud end in a semantic compression and multi-resolution representation vector mode, so that the transmission efficiency is improved; carrying out deep modeling and model performance monitoring at the cloud, and if degradation is detected, returning edge original data to update the model; and finally, generating a scheduling weight factor according to the prediction error distribution graph, and dynamically optimizing an edge cloud task allocation proportion. According to the invention, the problem of low resource allocation efficiency caused by lack of a dynamic scheduling mechanism based on prediction errors and confidence driving in existing mutual inductor test data edge cloud cooperative processing can be solved.

The invention discloses a production scheduling method and device based on digital twinning, equipment and a medium, and relates to the field of intelligent manufacturing, and the method comprises the steps: collecting production data, generating an initial production state, defining an initial production scheduling constraint condition, and constructing a digital twinning model; predicting the available capacity of the production resources through a digital twinborn model, identifying a risk period, and updating the twinborn state of the production resources; coupling the predicted production resource available capacity with the order set, generating a task priority queue and updating production scheduling constraint conditions; executing scheduling optimization and simulation detection conflicts in the digital twin model, if conflicts exist, executing local rearrangement based on conflict information until requirements are met, and generating an executable scheduling scheme; and issuing the executable scheduling scheme to a production and logistics link, monitoring execution deviation, rescheduling an affected task, completing a production task and updating the digital twin database. According to the invention, the real-time performance and accuracy of scheduling response are improved.

The invention discloses a front-end task scheduling and event processing method and device, equipment and a storage medium, and relates to the field of distributed systems and network technologies, and the front-end task scheduling and event processing method comprises the steps of capturing events triggered by all sub-elements; processing the captured events into standard task objects in a unified format; adding the standard task objects into a scheduling queue according to a priority sequence, and performing de-duplication processing and task merging processing on all the standard task objects in the scheduling queue to obtain a cleaned scheduling queue; all the standard task objects are sorted again according to the priority in the multi-task priority queue; according to the priorities of the standard task objects in the multi-task priority queue, dynamically adjusting the resource allocation of each standard task object in the multi-task priority queue in combination with the current load of the system; and executing the standard task object allocated with the resources by combining an asynchronous execution mode with a user feedback mechanism. According to the method, the problems of task conflict and resource competition in high-concurrency interaction can be effectively solved.

The invention discloses a code spraying identification optimal path planning method based on an intelligent optimization algorithm, and relates to the technical field of code spraying control, and the method comprises the steps: dispersing a code spraying region into a three-dimensional space, and constructing a dynamic environment model of marking parameter constraint; based on the code spraying task priority queue, a hybrid optimization algorithm is adopted to carry out collaborative optimization of marking parameters and path nodes, and an optimal path candidate set is generated; an obstacle avoidance model is constructed based on the optimal path candidate set, a collision-free path is generated, the pose of a nozzle is adjusted in combination with visual feedback, and an anti-interference code spraying action sequence is output; monitoring the operation feedback of the code spraying equipment, and carrying out the iterative updating of the marking parameter library and the algorithm rule. According to the method, the core advantages of multiple intelligent optimization algorithms are fused, a mixed collaborative optimization architecture is constructed, the limitation of a single algorithm in a complex scene is effectively broken through, and a traditional path planning algorithm is easily troubled by a local optimal solution and is difficult to meet multi-dimensional constraint conditions at the same time.

The invention provides a semantic fusion graphical software development management method and system. The method comprises the following steps: acquiring a multi-user interaction behavior data flow containing an operation track sequence and an interface control triggering time sequence; according to operation track sequences of different users and a time-space association mode of an interface control triggering time sequence, constructing a dynamic semantic map; inputting a hierarchical mapping relationship in the dynamic semantic map into a preset edge computing node, and generating a real-time computing power distribution scheme matched with a logic dependency chain in the dynamic semantic map; analyzing a mapping relationship between a control triggering timestamp and a control type in the space-time association mode, and generating an interface control rendering priority queue; the synchronous regulation and control interface control renders the priority queue and an operation instruction response sequence corresponding to the logic dependency chain, and generates a graphical software development process synchronous instruction set; according to the method, the hierarchy and the dependency chain are fused to form the dynamic semantic graph, the hierarchical operation structure and the logic dependency relationship are constructed, and the high efficiency and the consistency of the graphical development process are ensured.

The invention discloses a path planning method based on an improved artificial potential field method, and the method comprises the steps: carrying out the modeling of an environment through a grid method, and constructing a potential field map reflecting the distribution of obstacles; and searching an optimal path on the potential field map by adopting a minimum heap-first queue structure and combining a heuristic function considering the potential field gradient. For a local minimum trap problem, a new intermediate target point is selected near a trap point for re-searching by analyzing the potential field gradient change on a path. For special terrains such as narrow channels, a search strategy is dynamically adjusted to adapt to constraints. And the smoothness of the path is improved by normalizing and smoothing the potential field value of the node. In the path execution process, the potential field map can be dynamically updated according to real-time environment information, local re-planning is carried out on the affected path, and therefore dynamic obstacles can be effectively dealt with. According to the method, the intelligence, the safety and the adaptability of path planning are remarkably improved.

The invention relates to the technical field of intelligent dolls, and discloses a multi-modal data processing task scheduling optimization method and system for an intelligent doll, and the system comprises a multi-modal collection module which obtains voice, visual images and environment sensor data in real time; the task decomposition module analyzes the data into parallel tasks according to a preset mapping table, and the tasks are matched after feature extraction, image segmentation and timestamp alignment data fusion; the resource allocation module allocates an isolation container instance with a running environment and a resource quota according to a node load and a task demand; the scheduling execution module distributes tasks to edge nodes for execution according to the priority queues, monitors states and receives results; and the dynamic updating module periodically adjusts the queue weight coefficient according to the result. The method correspondingly executes each step. According to the scheme, the multi-modal task processing efficiency and the system stability of the intelligent doll are improved, and the method is suitable for related scheduling scenes.

The invention provides a hybrid expert model reasoning optimization method based on speculative preloading, and belongs to the field of artificial intelligence safety. The method aims at improving the reasoning performance of a hybrid expert model (MoE for short) in a resource limited scene, and solves the technical problems of high video memory occupation and bottleneck in calculation and data transmission of the hybrid expert model in a reasoning stage. According to the method, through hot expert identification, speculative preloading and pipeline execution, a fine-grained expert parameter preloading strategy and an asynchronous loading mechanism based on a priority queue are designed, and effective overlapping of a calculation process and a data transmission process is realized, so that the reasoning efficiency is improved. The method comprises the steps that expert parameters are divided into hot experts and non-hot experts, the hot experts are loaded to a computing device memory in advance, parameters of the next layer are asynchronously preloaded according to prediction for activation of experts of the future layer, and meanwhile computing of the current layer is executed. According to the method, resource optimization reasoning is realized on hybrid expert models with different scales and structures on the premise of not changing the structure and precision of the model.

The invention belongs to the technical field of industrial equipment inspection, and particularly discloses and provides an industrial equipment-oriented multi-terminal dynamic task collaborative inspection system, which comprises the steps of collecting equipment real-time operation state data and inspection terminal space position information, generating a dynamic priority queue, and flexibly adjusting the priority of an inspection task; direct communication connection is established between inspection terminals, relay nodes are established by detecting adjacent terminals, and autonomous reconstruction of inspection network communication is realized; a to-be-detected device is split into physical detection units which can be operated independently, and a task fragment combination containing a time window and a space range is allocated, so that multi-terminal cooperative work is realized; the running state data and the historical normal value range are compared in real time, the adjacent inspection terminals are automatically triggered to cooperate with a reinspection instruction, and it is ensured that data are accurate and reliable; and updating the three-dimensional state atlas of the equipment according to the detection result and the load state of the communication network, and synchronizing to all the inspection terminals in the connection state to ensure the consistency of the state views of the equipment.

The invention relates to the technical field of generator control, and particularly discloses an SVG static var generator control method and system, and the method comprises the steps: constructing a reactive-voltage sensitivity matrix containing a node sensitivity weight based on the real-time measurement data of a power grid, and obtaining a node sensitivity matrix; a dynamic sorting strategy of device priority grading is generated, a distributed compensation priority queue of all devices is obtained, and the devices comprise an SVG cluster and a photovoltaic inverter cluster; executing reactive power distribution under a capacity constraint condition based on the distributed compensation priority queue, and generating a target reactive power compensation amount with the capacity constraint condition; generating a hierarchical coordination control instruction of the SVG and the inverter according to the target reactive compensation amount, and obtaining a hierarchical control instruction set; executing the distributed cooperative compensation driven by the hierarchical control instruction set, and completing the dynamic regulation of the power grid voltage; the reactive power regulation capability of different devices in a power grid can be fully utilized, the accuracy and response speed of reactive power distribution are improved, and resource waste and excessive compensation are avoided.

The invention belongs to the technical field of path planning, and provides a dynamic path planning method, system and equipment based on multiple mobile robots and a medium based on the multiple mobile robots in order to solve the problems that the calculated amount is increased sharply and unexpected situations cannot be handled in the current path planning of the multiple mobile robots, and the independent path planning of each robot is completed by adopting mixed A *-epsillon. A dual-domain priority queue is introduced, child nodes are generated through node expansion, then a path is re-planned for the robot with constraints, and therefore a conflict-free path of the robot is generated; and performing control sequence sampling on each robot, taking a conflict-free space-time path as a reference path, obtaining an optimal control sequence according to the cost value of each predicted trajectory, and adjusting the path in real time through rolling optimization so as to cope with a dynamic obstacle. The method can actively avoid a dynamic obstacle or an obstacle which is not considered during planning.

The invention discloses a multi-queue scheduling method based on total and incremental reasoning overhead merging, which comprises the following steps of: firstly, calculating the total resource demand and memory overhead of a task according to the input length and the size of a predictor model in the process of performing total reasoning on a reasoning task; distributing tasks to queues with different priorities according to the total resource demand in combination with response time requirements and task importance; dynamically scheduling tasks according to a priority queue sequence, adjusting an execution sequence in combination with a system load, and preferentially executing tasks in a high-priority queue; distributing a time slice for each task, monitoring task execution time, reducing the priority of overtime tasks, moving the overtime tasks into a secondary priority queue, and triggering new task scheduling; and after preemptive scheduling, the key value cache of the latest scheduling task in the waiting queue is swapped from the accelerator memory to the CPU memory, and is swapped back to the accelerator memory before the task is re-executed.

The invention discloses a cloud server physical GPU acceleration capability release method and system, and belongs to the technical field of computer GPU capability release, and the method specifically comprises the steps: collecting GPU operation parameters in real time, carrying out the preprocessing, storing the GPU operation parameters in a shared state database, constructing a lightweight virtualization abstraction layer on a physical GPU, dividing the physical GPU into a plurality of logic units, and storing the logic units in a shared state database; the method comprises the following steps: establishing a physical GPU, constructing an isolation layer, utilizing a three-dimensional resource segmentation strategy, adopting a priority queue, a jump mechanism and a delay tolerance mechanism in a scheduling process, dynamically allocating logic unit resources of the physical GPU, accelerating release of GPU resources, and realizing unification and real-time updating of physical GPU state information among nodes of a cloud server through a distributed state synchronization mechanism; through dynamic task scheduling, a jumping mechanism, a delay tolerance mechanism and distributed state synchronization, efficient GPU resource management is realized, and GPU capability release is accelerated.

The invention discloses an aircraft real-time resource allocation method and system, relates to the technical field of aircraft control, and is used for solving the technical problem of conflict between resource limitation and demand caused by aircraft resource allocation, and the aircraft real-time resource allocation method comprises the following steps: constructing a task priority queue and a resource state matrix of an aircraft, defining a state space; defining an action space, and defining a reward function according to the dynamic resource allocation target; building a real-time control software environment of the aircraft; preparing training data of a double-depth Q network, and constructing a double-depth Q network structure; constructing an experience playback buffer area, carrying out dual-depth Q network training, and updating parameters of a main network; obtaining a value function between the state and the action of the aircraft, and enabling the aircraft to make an optimal resource allocation decision through the value function; dynamic planning is used to refine and allocate resources in real time, maximization of task income under the condition of meeting resource constraints is achieved, and real-time resource allocation of the aircraft is completed.

The invention provides a stream computing resource scheduling method and system based on a dynamic time window, and relates to the field of stream computing resource scheduling, and the method comprises the following steps: constructing a master-slave event time monitoring network to obtain a task feature vector, and establishing an adaptive time window scheduling mechanism to dynamically adjust the window size; and constructing a multi-level task priority queue, carrying out state transition by adopting a remote direct memory access mechanism of incremental transmission, and finally coordinating computing nodes through a distributed task arrangement protocol to complete resource switching. According to the method, computing resource competition can be effectively reduced, the flow computing task processing efficiency is improved, dynamic optimal configuration of resources is realized, and the system throughput is enhanced.

The invention discloses a high-frequency power data acquisition adaptive frequency and priority processing method and device based on resource load feedback, and a storage medium, and belongs to the technical field of high-frequency power data processing. The method comprises the following steps: acquiring resource load information of a distribution network side intelligent terminal node in real time; according to the real-time resource load information, judging whether a preset sampling period of the power monitoring data needs to be adjusted, if so, dynamically correcting through a double-layer fast and slow ring adjusting mechanism to obtain a final sampling period, and otherwise, maintaining an original period; acquiring power monitoring data acquired by the node in the corresponding sampling period; and on the basis of a preset priority queue grading rule, a tube queue algorithm is utilized to adopt a differential transmission strategy for the data, so that priority processing of different levels of data is realized. The method does not need to depend on a prediction model, and realizes acquisition link elastic control and key data delay guarantee through real-time quantification of node loads, dual-time-domain closed-loop adjustment of a sampling period and hierarchical queue management messages.

The invention relates to the technical field of intelligent monitoring, in particular to a multi-thread low-power-consumption intelligent monitoring system based on an AI processor. The method has the advantages that aiming at the problems of unbalanced computing power and power consumption, high multi-task processing delay and strong hardware dependence in the prior art, the NPU module of the RK3588 processor is combined with the INT8 quantitative model, so that the power consumption is lower than 10W under the 6TOPS computing power; a dynamic multi-thread scheduling mechanism is designed, parallel processing of more than eight paths of video streams is supported through binding of a priority queue and an NPU core, and end-to-end delay is compressed to be within 200 ms; a zero-copy video stream architecture is constructed, data transfer is eliminated through memory mapping, and preprocessing time consumption is reduced by 90%; an energy efficiency control module is integrated, the NPU voltage frequency is dynamically adjusted according to the load, and the energy efficiency ratio reaches 0.83 TOPS / W; space-time alignment of multi-model reasoning results is realized by adopting a frame ID synchronization technology, and the mismatching rate is lower than 0.1%.

The invention provides a multi-terminal collaborative AI model dynamic deployment method, which comprises the following steps that: a cloud terminal splits a complete AI model into lightweight sub-models according to a real-time state reported by edge equipment and AI model structure characteristics; calculating the sequence of the sub-models distributed to the edge devices through a priority queue algorithm; differential increment transmission is adopted to push difference parameters of the existing model of the edge device; the edge device triggers incremental learning based on an end side abnormal sample, and local model parameters are updated by adopting lightweight transfer learning; the cloud receives the encrypted local model parameters and the associated sample feature values, node parameters of the multiple edge devices are aggregated through federal learning, and an optimized global model is generated. According to the multi-terminal collaborative AI model dynamic deployment method, through a model fragmentation pushing technology and a side incremental learning mechanism, model dynamic adaptation and closed-loop optimization of resource awareness are realized, and the AI model deployment efficiency and scene adaptability in a complex industrial scene are remarkably improved.