304 results about "Network integration" patented technology

Disclosed in the present invention is a four-network integration architecture for an unmanned swarm system. The four-network integration architecture has the capabilities of heterogeneous platform resource pooling, intelligent dynamic computing power allocation, and timely decision planning, so as to maximize the overall benefit. The present invention focuses on abstracting and integrating independent submodules to form a mesh topology of a swarm. The present invention designs a four-network integration architecture for an unmanned swarm system, which comprises a computing power network, a perception network, a decision network and a communication network as core modules. The structure aims to achieve efficient cooperation of all parts in the swarm, thereby improving the overall performance and adaptability of the system. The system integrates environmental perception, a swarm network modeling component, a knowledge base and a resource pool, providing an intelligent environmental perception strategy and a network modeling strategy for the interior of the swarm. Therefore, the perception of environments, tasks and networks by nodes can be facilitated, thereby completing establishment of intelligent networks, so as to ensure the characteristics of the stability and flexibility of networks.

The invention discloses a container number tallying identification method, and relates to the technical field of container intelligent tallying. According to the technical key points, the method comprises the following steps: deploying a multi-modal sensing device, carrying out initial calibration, and pre-training an improved YOLOv7-tiny target detection model, a transformer-based cGAN character segmentation model and a lightweight CNN-transformer-capsule network integrated hybrid recognition model; through cooperative collection of multi-modal data of the multi-modal sensing device, the system can effectively integrate texture information of RGB images, a space structure of depth images and three-dimensional coordinates of point cloud data, the problem that traditional single visual recognition is affected by illumination, shielding and the like is solved, a self-calibration mechanism corrects space-time deviation of a sensor in real time, and the accuracy of recognition is improved. The system can still accurately extract container number characteristics in complex scenes such as strong light reflection, low illumination at night, character abrasion and the like.

The invention discloses an emergency broadcast message scheduling system, a processing method and a broadcast terminal, and belongs to the technical field of message scheduling. Emergency event data are collected, public events and regional events are divided according to an initial influence range, and SIP labels are constructed to generate broadcast messages; constructing a dynamic list based on a terminal identity and a network state, starting multi-network integration to execute a differential transmission strategy, and ensuring message delivery; public event preemption, regional event dynamic polling and priority jump triggering during event diffusion are realized through a multi-stage scheduling queue; volume coverage mapping is established by using multi-dimensional information of the terminal, the volume is dynamically calibrated in combination with regional characteristics, formats are adapted, and non-sensitive regulation and control are realized. The objective of the invention is to solve the problems of rigid multi-event scheduling, unstable heterogeneous network transmission, insufficient terminal adaptation capability and the like of a traditional system, and to improve emergency scene message scheduling efficiency, transmission reliability and coverage accuracy.

The invention provides a quadruped robot fault-tolerant control method and system based on residual learning, and the method comprises the steps: constructing an ontology mechanism model based on phase information, dividing the phases of a supporting stage and a swinging stage based on a diagonal gait, and designing foot end tracks through combining a Bezier curve and a sine curve; designing a six-dimensional reward function including speed tracking, posture balance, foot movement direction, energy consumption control, body contact constraint and foot end contact excitation; a data-driven model based on a heterogeneous actor-commentator architecture is constructed, an actor network integrates terrain information, ontology sensing data and damage parameter estimation values, a commentator network integrates privilege information for strategy evaluation, and network parameters are optimized based on a near-end strategy optimization algorithm; and on the basis of a residual learning thought, a final motion instruction is generated by coupling the correction output by the data driving model with the ontology mechanism model.

The invention discloses an insulator defect detection method and device, a medium and equipment, and relates to the technical field of power system equipment detection. The method comprises the following steps: integrating a star network StarNet into a YOLO11n network, adding a small target feature pyramid network into a neck network of the YOLO11n network, applying an ADown module into a backbone network and the neck network of the YOLO11n network, and replacing an IOU evaluation index of a loss function of the YOLO11n network with an NWD evaluation index to obtain an improved YOLO11n network; taking the training data set marked with the insulator state category as input, taking the insulator state category as output, training the improved YOLO11n network, and obtaining an insulator defect detection model; and inputting insulator image data acquired in real time into the insulator defect detection model to obtain a defect detection result of the insulator. According to the scheme, the defect detection accuracy of the insulator can be improved.

The invention provides a computing power resource scheduling method and system based on cloud network integration. Selecting a computing power task to be scheduled as a current scheduling task, and generating a node resource adaptation matrix based on the real-time load data, the cloud network topological relation between the nodes and the dynamic bandwidth data; then calculating the lowest scheduling cost of each node according to the matrix, and determining a dynamic adjustment factor when a resource allocation conflict occurs in the current scheduling task; based on the lowest scheduling cost, the computing power demand scale and the data transmission estimated overhead, calculating the final scheduling overhead for scheduling the current scheduling task to each node; and finally, allocating tasks to a target node according to the final scheduling overhead, updating a running task queue, if a conflict occurs, calling a dynamic adjustment factor to execute resource reallocation, and updating the queue after the reallocation succeeds. And circulating the process until all tasks are scheduled. According to the scheme, optimal scheduling of computing power resources in a cloud network convergence environment can be realized.

The invention discloses a battery health management system and method, and belongs to the technical field of battery management, the system comprises an implantable sensor network, an edge and cloud collaborative computing platform, a multi-modal prediction model and a dynamic optimization execution unit; the implantable sensor network integrates various sensors to collect multi-physical field parameters in a battery, and the energy collection unit supplies power through cross-shielding communication transmission. The edge and cloud collaboration platform processes data, trains a model and carries out block chain evidence storage; the multi-modal prediction model is fused with space-time double-flow Transform and causal reasoning, SOH and RUL are predicted, and thermal runaway is early warned; the dynamic optimization execution unit realizes efficient control through layered equalization and self-adaptive thermal management; the method comprises an initialization stage, an operation stage and a maintenance stage to form a closed loop. The method breaks through the traditional limitation, improves the safety, reliability and economy of the battery, and is suitable for electric vehicles, energy storage power stations and other scenes.

The invention relates to the technical field of computing network fusion, in particular to a green computing power intelligent scheduling optimization system for computing network fusion, which comprises a theft source sensing unit, a computing power distribution optimization unit, an energy consumption monitoring regulation and control unit and a task dynamic adaptation unit. A dynamic resource distribution graph is generated through multi-dimensional data collection, task allocation is optimized based on an improved fast transmission algorithm, and efficient and flexible task scheduling and energy consumption control are achieved in combination with hierarchical energy consumption management and a dynamic task adaptation strategy. According to the method, the computing power resource utilization efficiency can be remarkably improved, the energy consumption is greatly reduced while the service quality is ensured, and a brand new solution is provided for green computing in a computing network integration environment.

The invention provides a vehicle network integration cooperative regulation and control method and system. The method comprises the following steps that 1, the system collects access parameters of multiple types of charging facilities; 2, constructing a power distribution network safety evaluation model based on the data collected in the step 1; 3, determining a charging label of the accessed electric vehicle according to the product of the real-time SOC and the SOH; 4, comparing the three-phase voltage deviation, the current unbalance degree and the load growth rate of the power distribution network with the history in the same period to obtain an overload risk rate, and further calculating a power grid safety margin coefficient; 5, calculating a final safety evaluation result to truly reflect the vulnerability level of the current power grid; and step 6, constructing a digital twinborn model of the target area, performing simulation verification on the initial instruction, and collecting equipment state, vehicle response and power grid operation data after regulation and control. By applying the technical scheme, an efficient and collaborative vehicle network interaction system can be realized, and deep fusion and collaborative development of the energy and traffic fields can be promoted.

The invention relates to the technical field of safety monitoring, in particular to a multi-dimensional safety monitoring method and an early warning system of an integrated energy storage system. The method comprises the following steps: acquiring structural data of the energy storage system; performing energy storage unit analysis on the structure data of the energy storage system to generate energy storage unit data of the energy storage system; performing unit operation data monitoring on the energy storage unit data to obtain an energy storage unit operation data set; data monitoring network integration is carried out through the energy storage unit data and the energy storage unit operation data set of the energy storage system, an integrated energy storage system monitoring network is generated, and the integrated energy storage system monitoring network comprises a battery operation node, a flywheel operation node and a super capacitor operation node; according to the invention, multi-dimensional monitoring, real-time data analysis and artificial intelligence technologies are introduced, so that the comprehensiveness, real-time performance and fault prediction capability of safety monitoring of an existing integrated energy storage system are improved.

The invention relates to an energy accumulator durability test method and system, and the system comprises an extreme working condition simulation module which can achieve high-voltage, wide-temperature-range and high-frequency dynamic loading; the multi-sensor monitoring network integrates a fiber bragg grating strain sensor, a three-axis acceleration sensor and an acoustic emission sensor to realize collaborative monitoring of strain, vibration and microcracks; the intelligent control and energy recovery module is combined with a fuzzy PID and reinforcement learning algorithm, a magnetorheological fluid generator and a super capacitor bank, so that the control precision and the energy utilization rate are improved; and the data analysis and life evaluation platform accurately predicts the residual life by using an improved Bayesian network model and a real-time health index calculation module. According to the test method, a loading mode is dynamically adjusted through a real-time health index, an energy recovery strategy is optimized, and life early warning is triggered. The invention breaks through the limitation of the traditional test and provides an integrated solution.

Embodiments manage data in a file system over a network. A plurality of file system operations in the file system may be executed based on a file system client action or a file system administrative action such that the file system may be integrated with a network. Characteristics of the plurality network components in the network infrastructure that may be associated with the file system may be determined. Tests may be generated based on the characteristics of the network components such that the tests may be executed to evaluate the network components. Results of the tests may be employed to perform further actions, including determining non-compliant network components based on the results; modifying the network infrastructure based on the non-compliant network components such that one or more of file system operations are modified based on the non-compliant network devices; executing the modified file system operations on the modified network infrastructure.

The invention discloses a power distribution network reconstruction method and system based on topology security constraint and integrated reinforcement learning. The method comprises the steps of power grid environment modeling including system state quantity design, reward function design, power grid constraint embedding and system power flow calculation; constructing a topology security layer, and correcting a topology reconstruction action by using a topology mask through a detection algorithm; pre-judging different actions by using an integrated action network, a parameter doping mechanism and a network deletion mechanism; building a reinforcement learning training framework to train the initial strategy network, and obtaining network parameters; and obtaining a scheduling strategy of power distribution network reconstruction according to the state data of the system by using the trained strategy network. According to the method, a topology security layer is superposed on the basis of reinforcement learning, and mechanisms such as action network integration are introduced, so that the problem of lack of security of reinforcement learning model output strategies is solved, the volatility of the parameter training process is relieved, and the reconstruction scheduling strategy of the power distribution network can be efficiently obtained in real time.

The invention discloses a cloth surface flaw detection method and device based on texture perception and anomaly detection. The method comprises the following steps: acquiring a surface image of detected cloth; inputting the surface image of the detected cloth into a deep learning network model; a multi-scale feature map is extracted through the backbone network; processing the feature map through the texture perception feature extraction module so as to fuse cross-channel and cross-space texture information; integrating anomaly detection branches through the check network, and outputting feature maps of different scales; performing frequency domain enhancement and spatial domain feature extraction operation and fusion on the features through the adaptive frequency domain convolution module; and outputting a detection result through the YoloHead detection head so as to judge whether the cloth has flaws or not. According to the method, the texture feature information in the cloth image can be effectively utilized, and the detection accuracy and robustness of the cloth surface flaws and the recognition capability of unknown flaws are improved.

The invention provides a lightweight medical image segmentation network (ABUNet) based on asymmetry and an implementation method thereof, and the method comprises the following steps: S1, in a coding stage, proposing a feature subtraction convolution block (FSCB), and implementing O (C2 / N)-level parameter compression (N is a group number) by using channel feature difference operation; in a lightweight scene, the FSCB can effectively reduce feature redundancy, directly highlights key features of a lesion area, and is superior to traditional feature operation based on addition and multiplication; s2, in a decoding stage, a feature addition convolution block (FACB) is designed, a multi-branch feature fusion mechanism is adopted, and the alignment precision of different feature representations is improved under the condition that the calculation cost is not increased; and S3, in a bridging stage, a multi-scale deep convolutional block (MSDB) is constructed, and the multi-scale context modeling capability of the model is remarkably enhanced by utilizing heterogeneous kernel parallel computing, so that more accurate lesion feature extraction is realized. And S4, in a network integration stage, an FSCB module is integrated into an encoder part of a U-shaped architecture, an FACB module is integrated into a decoder part, and an MSDB module is used for processing grouping characteristics in a bridging module to construct an asymmetric model ABUNet. The asymmetric architecture overcomes the limitation of symmetry of a traditional encoder-decoder, and effectively balances high segmentation precision and calculation efficiency.

The invention discloses a cross-individual EEG emotion recognition method based on course learning and multi-source domain adaptation, and the method comprises the steps: constructing a multi-modal feature extraction network, and integrating feature extraction modules of time sequence, frequency domain, brain network connectivity and the like, so as to fully mine and fuse multi-dimensional emotion related information. Aiming at challenges of large individual difference, inconsistent data distribution and the like in cross-subject emotion recognition, a multi-source domain adaptation mechanism is introduced, and the discrimination ability and migration performance of a model on a target domain are improved by effectively aligning feature distribution of a source domain and a target domain. Besides, in order to relieve the problems of unstable convergence and local optimization caused by difficult samples in the initial training stage of the model, a course learning strategy is introduced, target domain data is guided to participate in training from shallow to deep according to the sample difficulty, and therefore the convergence, generalization ability and robustness of the model are remarkably enhanced, and the training efficiency is improved. And finally, the overall performance of cross-subject EEG emotion recognition is effectively improved.

The invention relates to the technical field of neural network multilayer perception, in particular to a sensitive data auditing method based on neural network multilayer perception, which comprises the steps of data collection, data preprocessing, feature selection, multilayer perception neural network integration, model training, real-time data stream processing, model evaluation, result analysis and privacy protection enhancement. According to the invention, through the deep learning ability of the multi-layer perceptron (MLP) neural network, the perception and extraction ability of the data features is significantly improved. Due to the introduction of a real-time data stream processing technology, the system can respond to sensitive data leakage in time, and the security risk is effectively reduced. The self-adaptive learning mechanism ensures that the model can continuously adapt to novel threats, and the efficient detection capability is kept. The differential privacy technology provides powerful privacy protection in the data processing and model training process, and the risk of user privacy disclosure is avoided.

The invention discloses an emergency linkage control system for oil depot fire hazard real-time monitoring, and relates to the technical field of safety monitoring, the emergency linkage control system comprises an emergency linkage control center, the emergency linkage control center is in communication connection with the following modules: a multi-source data fusion sensing module used for constructing a multi-mode sensor network, and collecting multi-source heterogeneous data in real time, carrying out fusion analysis, and generating a hidden danger dynamic distribution diagram. Real-time high-frequency acquisition of oil depot environment, equipment and disaster evolution data is realized by constructing a multi-mode sensor network and integrating various sensors, data noise is eliminated and a hidden danger dynamic distribution diagram is generated in combination with a spatio-temporal data fusion engine, so that the monitoring data covers the whole area of the oil depot and the time resolution reaches a second level; compared with a traditional threshold triggering mode, the system can dynamically capture multi-source heterogeneous data of the key area of the oil depot, discover the hidden danger evolution trend in advance, shorten the disaster early warning time and remarkably improve the active suppression capability.

The invention belongs to the technical field of digital terrain analysis and artificial intelligence crossing, and particularly discloses an automatic identification method and system for basins in DEM data based on a neural network. The method comprises the following steps: adaptively determining a topographic relief amplitude analysis window based on a mean value point change method, and extracting an initial plain area by combining elevation, gradient and topographic relief amplitude; noise is eliminated through morphological optimization (hole filling and edge smoothing), a buffer area is constructed, and key terrain factors such as elevation and gradient change rate are screened to generate multi-channel feature data; dividing landform types by using ISODATA dynamic clustering, and outputting binary basin data; an improved ResUNet + + network (integrating multiple encoders, a cavity space pyramid and a channel attention mechanism) is adopted for end-to-end training, and refined segmentation of a complex boundary is achieved. The method supports cross-regional generalization testing, can realize high-precision basin identification in drought and moist landform scenes, and can be widely applied to the fields of resource exploration, ecological protection and disaster prevention and control.

The invention belongs to the technical field of solid waste treatment, and particularly relates to a reprocessed plastic blow molding product quality stability self-adaptive adjusting system. And the blow molding process parameters are optimized by using neural network integration, firefly algorithm optimized fuzzy-PID control, deep reinforcement learning, Bayesian optimization and a chaos optimization algorithm. According to the algorithms, for enterprises with different scales and requirements, the product quality is effectively improved, the production efficiency is improved, the defective rate and the energy consumption are reduced, and diversified quality optimization strategies are provided for the regenerated plastic blow molding industry.

The invention belongs to the technical field of computing power networks, and discloses a self-adaptive heterogeneous computing power network and an operation method thereof. The network comprises a computing network infrastructure layer which is used as the bottommost layer of the adaptive heterogeneous computing power network and comprises computing power resources and network resources; the resource abstraction layer is used for abstractly packaging computing power resources and network services provided by the computing network infrastructure layer; the computing network integrated layer is used for preprocessing and joining resource objects; the computing network brain layer is used for realizing collaborative scheduling of resources of the whole network; and the heterogeneous computing power network service base layer and the user service layer are used for integrating bottom computing power resources, network resources, intelligent arrangement and whole-network collaborative scheduling services into standardized services. According to the method, the existing heterogeneous computing power resources are constructed into the self-adaptive heterogeneous computing power network, the computing power bottleneck of a single computing power resource is broken through, meanwhile, a unified and more powerful computing power service is provided for a user, and the user does not need to care about the specificity of each computing power resource.

The invention provides a mining 5G cloud network convergence communication system and method, and the system comprises a cloud network convergence platform, a 5G private network core network, a transmission network, a 5G wireless network, and an intelligent terminal. The intelligent terminal is used for collecting information and transmitting the information to the cloud network integration platform through the 5G wireless network, the transmission network and the 5G private network core network in sequence; the cloud network integration platform divides mining 5G communication services into multiple service types, and each service type corresponds to one or more intelligent terminals; the cloud network convergence platform allocates computing power resources to the mining 5G communication services based on a computing power resource allocation strategy according to the service types of the mining 5G communication services; the 5G private network core network and the transmission network provide customized network slices for the mining 5G communication service according to a set slice strategy and a service type; and intelligent resource distribution and computing power distribution of the mining communication system are realized.

The PIO communication device comprises a protocol conversion module and a PIO transaction processing module, the protocol conversion module and the PIO transaction processing module are connected with each other, the protocol conversion module is connected with a computing component connected to a network-on-chip NoC through an AXI interface, and when the computing component initiates a PIO read-write request, the PIO transaction processing module sends the PIO read-write request to the computing component; the protocol conversion module performs conversion of read-write requests between an AXI protocol and a PIO protocol and mapping of response states, and the PIO transaction processing module initiates register access and descriptor writing operation to the network protocol processing component by analyzing the converted PIO read-write requests and receives register read response data returned by the network protocol processing component. According to the invention, a decoupling design of an address channel, a data channel and a response channel is adopted, conversion of a PIO read-write transaction between an AXI interface and a network protocol interface is completed, and high-bandwidth and low-delay non-blocking PIO data transmission and processing are realized.

The invention discloses a milling process control method based on reinforcement learning, which comprises the following steps: collecting and processing multi-source process data in real time, normalizing and extracting features, and forming state feature vectors; the state feature vectors are input into a DSAC-T reinforcement learning controller, the opening degree of a feeding valve and the rotation speed adjustment of a main shaft are used as action spaces, a reinforcement learning strategy network is constructed, and a process drift online detection sub-module is integrated; training a controller to maximize the rice yield, minimize the broken rice rate and energy consumption, and combining the process safety index to obtain an optimal strategy; deploying the trained controller in an actual scene, and outputting an adjustment instruction to implement closed-loop control; when the actual rice grain size or broken rice rate exceeds the model prediction deviation, online incremental learning is automatically triggered, and control parameters are optimized; the intelligent, self-adaptive and safe control of the rice milling process is realized. According to the method, adaptive optimization control of the rice milling process is realized, and the method has efficient response capability to dynamic disturbance and abnormal states.

The invention discloses a constructed network type energy storage converter and a control method thereof, and a constructed network integration system. The constructed network integration system comprises a transformer which is provided with a first connecting end, a second connecting end and a third connecting end; the direct current side of the first inverter is connected with the new energy generator, and the alternating current side of the first inverter is connected with the first connecting end, one end of the first switch and one end of the second switch; the direct current side of the second inverter is connected with the energy storage unit, and the alternating current side of the second inverter is connected with one end of the third switch and the other end of the second switch; the other end of the third switch is connected with the second connecting end; the third connecting end is connected with the other end of the first switch and the power grid; and the controller is used for controlling on or off of the first switch, the second switch and the third switch, and is also used for switching a control strategy of the second inverter. The weak network adaptability of the network-following converter and the fault ride-through capability of network-following integration can be effectively improved.

The invention discloses a self-adaptive distributed network integration system and method for virtual and real unmanned aerial vehicle cooperation, and relates to the technical field of unmanned aerial vehicle control, the system comprises a central control plane and at least one node agent, the central control plane comprises a global identity manager configured and distributed with globally unique network identity information, recording and recovering the network identity information; the network topology and state manager is configured to generate a virtual and real network state view, responds to query and issues a network simulation instruction to the node agent; the distributed network coordinator is configured to decide and coordinate the node agents to construct a cross-host logic two-layer network domain; the security policy manager is configured to define and store a network access control policy of the virtual unmanned aerial vehicle; the adaptive distributed network integration system and method aim to realize physical presentation of the virtual unmanned aerial vehicle on the network level, so that the virtual unmanned aerial vehicle can directly communicate with and collaboratively work with a real unmanned aerial vehicle in the same network plane.

The invention discloses an all-network public network communication network networking fusion method and system based on satellite communication. The method and system are suitable for emergency communication guarantee, remote area communication coverage and other scenes. The system comprises a communication access module, a data convergence module, a three-network convergence gateway, an operator core network and a unified management and scheduling module, wherein the unified management and scheduling module is used for subdividing a core network link monitoring subsystem and a convergence site management subsystem. According to the system, by means of a three-network integration gateway, a single pico-base station device can provide signal coverage of three operators at the same time, a communication link is connected through a private network and encrypted, and safety is guaranteed. The method is based on the system and comprises the steps of communication access, data aggregation, data processing and management, service processing, management and scheduling and the like. The multi-operator network selection requirement can be met, the network security and stability are enhanced, resource centralized management and monitoring scheduling are achieved, and the method has important application value in emergency rescue, remote area communication coverage, large-scale activity communication guarantee and other scenes.

The invention relates to the technical field of network resource allocation and management, and discloses a dynamic demand-based computing network integration multilayer scheduling optimization method, which comprises the following steps of: S1, collecting and preprocessing related data of computing power and network resources; s2, according to the pre-processed related data, constructing a network resource model for converting a discrete computing power and network resource allocation problem into a continuous optimization problem; and S3, applying the constructed network resource model to an actual computing network fusion system, solving a computing network fusion multi-layer scheduling optimization problem through an optimized genetic algorithm in combination with multi-level computing power and a network resource dynamic scheduling strategy, and accurately sensing and analyzing task requirements and real-time load states. And dynamic scheduling and optimal distribution of computing power and network resources are realized. According to the method, the resource utilization efficiency is improved, resource waste is reduced, and the task execution efficiency is optimized.

The invention relates to a tunneling action generation method and system based on time-space depth fusion multi-task prediction in the technical field of shield engineering data processing, and the method and system integrate local and global time sequence information through a dynamic depth fusion network, achieve the fusion of multiple time-space scales, improve the perception capability of complex working conditions, and improve the efficiency of shield engineering data processing. Meanwhile, future state prediction and control candidates are output, action fusion is carried out on a strategy layer, performance loss caused by prediction-control splitting is reduced, prediction-control integration is achieved, an uncertainty head is introduced, explicit constraint is carried out on a loss function and strategy fusion layer, the risk under stratum sudden change or sensing noise is effectively restrained, and the prediction-control performance is improved. Uncertainty constraint security is realized; on the basis of experience playback, weight self-adaption and noise removal, online self-adaption updating requirements of different stratums and tunneling stages are met, online self-adaption updating can be achieved, and the action generation capacity of tunneling stability control under the complex stratums and noise conditions is remarkably improved.

The invention discloses an intelligent operation and maintenance method and system for cloud network integration, and relates to the technical field of communication, and the method comprises the following steps: extracting a probability distribution density according to the reason and frequency of a fault of a historical sample; selecting similar distribution based on the probability distribution density, predicting the future occurrence probability of the fault, and sending the fault sample data for simulation when the future occurrence probability of the fault exceeds a threshold value to obtain a first evaluation verification result; performing entity fault detection based on the current network environment to obtain a second evaluation verification result; and integrating the first evaluation verification result and the second evaluation verification result to determine a fault reason. According to the invention, the technical problem that fault prediction and fault evaluation in a cloud network cannot be realized in the prior art can be solved.