18 results about "Cloud data center" patented technology

The application discloses a kind of deep neural network gas identification method and system based on internet of things electronic nose, gas sample response signal obtained by gas sensor array collection is sent to cloud data center by terminal equipment with sensor array and internet of things module.In the pre-processing of data to microcontroller MCU, gram angle field transformation is carried out, so that the two-dimensional sensor response data received by host computer becomes three-dimensional data that can be input into convolutional neural network after dimensionality increasing, and the gas label is output through classifier, which can represent gas category and concentration grade, realizes the real-time, anti-interference and high detection accuracy of remote monitoring.Gives play to the advantages of strong feature extraction ability, fast model convergence and high recognition accuracy of convolutional neural network.It can be widely applied in environmental detection, industrial production, medical treatment and safety fields.

This invention relates to the field of video recognition technology, and discloses a method and system for real-time anomaly recognition of security video streams based on edge computing. The method includes: acquiring real-time video streams from monitoring devices at edge nodes, decoding and standardizing them to obtain a sequence of image frames of the target area; extracting features from each frame of the sequence and performing preliminary anomaly detection to generate a preliminary anomaly detection result; when the result indicates the presence of a potential anomaly, triggering deep behavioral analysis, performing spatiotemporal feature fusion analysis on the associated continuous image frame sequence to obtain a behavioral semantic description vector of the target area; calculating the deviation of this vector from the normal behavioral pattern cluster at the edge nodes to determine the anomaly type label and confidence level; generating and issuing a graded alarm according to a preset alarm strategy, and simultaneously packaging the anomaly label, associated image frame summary, and alarm signal into a structured log and uploading it to a cloud data center. This invention can improve the efficiency of real-time anomaly recognition of security video streams.

This invention discloses an adaptive cloud-edge-device collaborative task offloading method based on multi-agent reinforcement learning, belonging to the field of cloud-edge collaborative task offloading technology. The invention includes: constructing a three-layer cloud-edge-device collaborative model; the end user optimizing the selected collaborative edge devices and the amount of offloaded tasks, with the edge servers including fixed edge servers and drone servers; the drone servers optimizing flight trajectories based on cloud-based scheduling locations; the cloud data center optimizing drone locations based on load conditions; modeling the mobile device energy consumption and task completion latency during the cloud-edge-device collaborative task offloading process as a comprehensive system cost optimization problem, further transforming it into a Markov decision process; and designing a multi-agent reinforcement learning algorithm to optimize the task offloading strategy. Under the premise of satisfying various constraints, this invention can effectively improve task execution efficiency while minimizing mobile device energy consumption, especially improving the service duration and quality of drones.

This invention relates to the field of edge-cloud and discloses a task offloading method for a three-layer IIoT system based on an improved multi-objective evolutionary algorithm. The method includes constructing a three-layer collaborative system model comprising a cloud data center, at least one edge server, and at least one industrial IoT device. The edge server is communicatively connected to both the cloud data center and the industrial IoT device. The method involves constructing a task model, a communication model, and a computation model; building a multi-objective optimization model that minimizes latency and energy consumption, with the average completion time of the industrial application and the average energy consumption of the terminal as the objective functions; initializing weight vectors and determining the neighborhood; initializing a mixed population; adaptively adjusting the weight vectors; encoding individuals and performing genetic operations; and outputting an offloading decision set. The advantages of this invention are that it can effectively reduce terminal energy consumption and edge node load while meeting the latency constraints of industrial applications, thereby enhancing computational efficiency and reliability.

The present application provides a container scheduling method and apparatus in a cloud environment, an electronic device, and a storage medium. By means of the present application, a sub-cluster manager and a resource cluster object are introduced into a sub-cluster, and a primary cluster manager is introduced into a primary cluster. The sub-cluster manager, the resource cluster object, and the primary cluster manager cooperate with one another to implement real-time resource sensing and resource balancing capabilities of each sub-cluster in a cloud data center in a federation scenario, thereby enhancing the real-time capability of detecting whether a container is normal / available, and ensuring that the container can be immediately scheduled to another node or cluster when an actually required system resource cannot be obtained by the container, thereby reducing the duration for which the container is unavailable, avoiding affecting the quality of an outward cloud service provided by the container, and improving the availability of the containers in each sub-cluster in the cloud data center in the federation scenario and the comprehensive resource utilization of the sub-clusters.

The application discloses a fracturing working condition monitoring method and system based on a virtual serial port data interception technology, which can efficiently collect operation parameters and fracturing construction parameters of field fracturing equipment, and ensure data compatibility through protocol conversion, and then safely transmit the parameters to a fracturing instrument vehicle computer. By starting virtual serial port setting software, seamless docking of physical serial ports and virtual serial ports is realized, which not only improves the flexibility of data reading, but also ensures the integrity and accuracy of data. After being buffered and checked on the instrument vehicle computer, the data is fed back to the data acquisition software of the fracturing vehicle in real time, and the other way is efficiently sent to the designated serial port through the virtual serial port, realizing the bidirectional flow of data. The read data is preprocessed, the key parameters are extracted, and are transmitted to the oilfield cloud data center in real time, providing solid data support for remote monitoring and intelligent analysis, thereby significantly improving the efficiency and intelligent level of fracturing construction.

This invention relates to the field of industrial internet and chemical equipment operation and maintenance technology, specifically to a remote intelligent operation and maintenance system for chemical utility equipment based on the Internet of Things (IoT). The system includes a field layer for the chemical utility equipment, a network transmission layer, a cloud data center layer, and a user terminal layer. The network transmission layer constructs a hybrid wireless communication network architecture incorporating 5G communication technology and low-power wide-area network (NB-IoT) technology. The cloud data center layer includes a big data storage server and a big data analysis server. The big data analysis server is equipped with a data analysis model based on deep learning, including convolutional neural network (CNN) and recurrent neural network (RNN) models. This invention enables early fault warning and accurate diagnosis, triggering tiered warnings when the fault probability exceeds 70%, assisting maintenance personnel in proactive handling, reducing manual intervention while improving equipment operational stability and energy efficiency.

This application discloses an anomaly detection device and method, which can automate the anomaly detection of optical modules in physical servers. The device can be installed on a rack in a cloud data center. Multiple physical servers can be placed in the hollow portion of the rack, and the optical modules of these physical servers can face the opening in the hollow portion of the rack. The anomaly detection device includes a moving part and a detector. The detector, supported by the moving part, can move freely at the opening in the hollow portion of the rack to face the optical modules of the multiple physical servers. During movement, the detector can send ultrasonic waves to the optical modules of the multiple physical servers. The ultrasonic waves are reflected at the optical modules of the multiple physical servers, and the reflected ultrasonic waves are received by the detector. Based on this, the detector can determine whether there is an anomaly in the optical modules of the multiple physical servers.

The application discloses a data resource contention risk prediction and optimization method based on LSTM and attention mechanism, and particularly relates to the technical field of computing resource management. First, the multi-dimensional performance index sequence of a cloud data center is collected, and the shared resource affinity of fusion time decay and resource access frequency is calculated as an auxiliary feature to construct an enhanced input for training an LSTM network. The attention mechanism fused with a contention diffusion speed factor is used to dynamically weight the key time slices and quantify the risk driving. The resource contribution mapping of the introduction of the causal test is used to determine the dominant contention resource, and the precise adaptive optimization action, such as virtual machine migration or path partition adjustment, is triggered accordingly, thereby improving the accuracy and interpretability of the resource contention risk prediction, realizing the closed-loop linkage from prediction to optimization, and solving the problems of difficult performance degradation prediction, inaccurate root location and low optimization efficiency caused by the implicit contention of micro-architecture level resources in the cloud data center.

This invention discloses a data-augmented anomaly localization method for IoT-sensing cloud data centers, relating to the field of cloud data. The method includes: S1 acquiring a training dataset; S2 constructing an anomaly detection model; S3 importing the training dataset into the anomaly detection model for optimization training; S4 acquiring the raw data from the cloud data center in real time and importing it into the optimized anomaly detection model to obtain predicted values ​​for the raw data; and S5 determining anomalies in the raw data of the cloud data center based on the predicted values ​​and the true values ​​of the raw data. The method utilizes data augmentation-neural transformation and convolutional long short-term memory networks to mine different aspects of the temporal and spatial characteristics of multivariate time series, increasing the amount of training data and reducing false positives caused by non-stationary and nonlinear real-time IoT data. Furthermore, it employs an attention-based autoregressive LSTM network to reduce the dimensionality of the data and extract features again, fully acquiring information beneficial to prediction from the data and improving the robustness of the model's anomaly detection.

Embodiments of the present application relate to a cloud resource pool energy-saving scheduling method, electronic equipment and products. The method monitors multi-dimensional cluster resource utilization rate indicators and the number of running state host computers, establishes a target baseline to define the reasonable interval of each indicator, when all hibernation determination conditions are met, performs a hibernation operation based on the minimum number of hibernation host computers required to return to the baseline of each indicator, so that the resource utilization rate and the number of host computers return to the baseline; when any wake-up condition is met, perform a wake-up operation based on the maximum number of wake-up host computers required to return to the baseline of each indicator, so that the resource utilization rate returns to the baseline. The scheme realizes multi-index collaborative decision and dynamic quantity calculation, overcomes the defects of single-index or fixed threshold method, can accurately balance resource utilization rate and energy consumption on the premise of ensuring business stability, avoids excessive hibernation affecting services or insufficient wake-up leading to performance degradation, and significantly improves the energy efficiency and automation management level of cloud data centers.

1. The name of the design product: graphical user interface for environmental and energy consumption data of electronic equipment (smart five constant cloud data center). 2. The use of the design product: for running programs, displaying environmental and energy consumption data information and communication. 3. The design points of the design product: in the content of the graphical user interface. 4. The picture or photo that best shows the design points: design 1 front view. 5. Design 1 is designated as the basic design. 6. The use of the graphical user interface: the graphical user interface is used to view environmental parameters such as temperature and humidity, PM2.5, carbon dioxide, noise, etc. The interface is used to display and count power, water flow, carbon emission energy consumption parameters. The interface is used to display five constant system data summary, fault alarm and equipment running state information. The interface is used for human-computer interaction of environmental and energy consumption data information. 7. The human-computer interaction mode of the graphical user interface: view environmental and energy consumption data information parameters through the graphical user interface.

A crop yield prediction device and method based on in-situ monitoring, relating to the field of crop yield prediction technology, includes a local data processing terminal and a cloud data center. The local data processing terminal is used for wireless data transmission with in-situ monitoring sensors and field weather stations, and for preprocessing data collected from these sensors and stations before uploading it to the cloud data center. The cloud data center embeds a crop yield prediction model. This invention employs core mechanisms such as multi-depth monitoring, water-fertilizer coupling, and nutrient weight allocation, demonstrating good predictive performance in practical applications.

This invention relates to the field of 6G communication technology, and in particular to a 6G satellite-to-ground computing power network offloading method based on lightweight blockchain and hierarchical reinforcement learning. The method includes: S1, constructing a computing power network model based on user local devices, low Earth orbit satellites, and remote cloud data centers; S2, calculating the latency and energy consumption of different communication paths based on the user's offloading decision; S3, establishing a latency model and an energy consumption model by combining the offloading ratio, latency, energy consumption, and blockchain system overhead; S4, solving for the optimal solution of the global optimization objective function to generate the optimal offloading action for each user. This invention solves the technical problems of insufficient collaboration of computing power resources in satellite-to-ground computing power networks, poor adaptability of consensus mechanisms, and lack of security systems, achieving the goal of optimizing the utilization efficiency of computing power resources across the entire domain and adapting to the dynamic and real-time requirements of satellite-to-ground networks.

This invention discloses a network and information security defense method and system based on multi-dimensional threat perception, relating to the field of network and information security technology. This invention obtains mirrored traffic data from key network nodes, separates transport layer security protocol handshake messages and application data messages, extracts unencrypted handshake metadata from client and server greeting messages using a first perception channel, generates a multi-dimensional static fingerprint vector, extracts the application layer payload length sequence using a second perception channel, calculates the packet length state transition probability distribution as a dynamic behavior feature vector, and jointly encodes the static fingerprint vector and dynamic behavior feature vector, inputting it into a shallow graph convolutional network to output a malicious confidence score. When the score exceeds a threshold, session-level blocking is executed. This achieves high-precision detection of encrypted malicious traffic without decrypting the traffic, and is suitable for encrypted traffic security protection scenarios at enterprise network boundaries, cloud data centers, and critical information infrastructure.

This invention discloses an anomaly detection method and system, relating to the field of cloud computing technology. The method includes processing collected time-series monitoring data into data processing objects with unique object identifiers according to data detection granularity; performing time-frequency analysis on the data processing objects to generate feature fingerprints with write-prohibited and modification-prohibited attributes based on the analysis results; using an anomaly recognition model to identify anomalies in the feature fingerprints, obtaining a first anomaly probability value and an uncertainty coefficient; if, based on the uncertainty coefficient and the first anomaly probability value, it is determined that the model identification result should not be adopted, a central controller determines a second anomaly probability value by retrieving and analyzing historical data, and combines this second anomaly probability value with the first anomaly probability value to determine the final anomaly detection result. This invention can solve the problem that related technologies cannot meet users' anomaly detection needs for cloud data centers, achieving a globally optimal balance between real-time anomaly detection, resource overhead, and diagnostic accuracy.