519 results about "Intelligent computing" patented technology

The invention discloses an earth and rockfill dam seepage-deformation early warning method and system based on time-space combined anomaly, and belongs to the field of dam body safety data research. The method comprises the following steps: constructing a spatio-temporal topological graph based on an engineering coordinate system, integrating multi-dimensional data by nodes, and constructing a dynamic adjacency matrix according to spatial distance and seepage relevance; extracting features by using a space-time diagram convolutional network, a self-loop mechanism and cross-layer attention; and executing dual-drive early warning through standard threshold preliminary screening, multi-scale LSTM prediction and a time decay evidence theory. The system comprises a sensor network and an intelligent computing module, and the intelligent computing module has adaptive modeling and visualization functions. According to the scheme, seepage-deformation space-time correlation quantitative analysis is achieved, the hysteresis effect is captured, the threshold value is dynamically corrected, multi-source evidences are fused, the early warning timeliness and accuracy are improved, and the risk of false alarm and missing alarm is reduced.

The invention discloses an embedded AI intelligent computing power architecture method, and particularly relates to the technical field of artificial intelligence processing architecture. Collecting a resource state parameter set R of the embedded device; obtaining a to-be-executed AI task set T, wherein each task comprises model complexity, real-time requirements, expected response duration and priority; constructing a computing power resource allocation evaluation function F based on R and T, and outputting a task scheduling priority score; allocating tasks to the embedded AI module according to the allocation scheme and executing reasoning; the resource state is dynamically monitored in the task running process, and if it is predicted that resources are about to be overloaded, a scheduling function F is triggered to reconstruct a resource allocation scheme; performing iterative optimization on weight parameters in the function F based on task history feedback; by means of the method and device, optimal adaptation of multi-task concurrent scheduling can be achieved under the condition that resources are limited, the computing power resource utilization rate, the response efficiency and the system stability are improved, and the method and device are suitable for various edge side AI scenes.

The invention relates to the technical field of industrial communication and Internet of Things, in particular to an IO-Link gateway system and method for heterogeneous protocol unification and edge enabling. Comprising a hardware layer, a protocol adaptation layer, a data processing layer, an intelligent computing layer and a communication interaction layer which are connected in sequence to achieve standardized conversion and edge intelligent processing of heterogeneous industrial data, and data time sequence consistency and system operation stability are guaranteed in the whole process. According to the invention, a plurality of heterogeneous industrial protocols are downwards and uniformly mapped to a technical path of an IO-Link international standard, and heterogeneous data are converted into a standard IO-Link object dictionary at a gateway level, so that field devices of different manufacturers and different protocols can carry out data interaction with a uniform and semantic-rich information model; therefore, plug-and-play of data is realized in the equipment access layer, and a solid foundation is laid for constructing an integrated industrial Internet of Things system.

The invention discloses an intelligent computing center data security protection method and system based on a computer network, and relates to the technical field of network data security protection. Comprising a data classification identification module, a data encryption and fragmentation module, a distributed storage management module, a network attack detection and defense module, a data integrity verification and recovery module, a security policy optimization module, a security situation awareness module and a central control module. The distributed storage management module records storage positions, hash values and access permissions of data fragments based on the block chain technology, the distributed storage management module dynamically adjusts the access permissions of storage nodes, node access is only allowed to be achieved through multi-factor authentication of fingerprints and digital certificates, the data streams are fragmented and then encrypted, and therefore the data streams are encrypted, the data streams are encrypted, and the access permission of the storage nodes is dynamically adjusted. And the data is obtained by reading the storage position, the hash value and the access permission, and the access permission of the storage node is dynamically adjusted, so that the data security can be ensured while the data can be conveniently accessed.

The invention relates to a data loading method and device, a storage medium and equipment, the method is used in an intelligent calculation data processing system comprising a host, a CSD and a SmartNIC, and a PCIe P2P data path is established between the CSD and the SmartNIC; the method comprises the steps that a host generates collaborative descriptors and respectively issues the collaborative descriptors to a CSD and a SmartNIC; the CSD reads source data from a storage medium according to source address information in the collaborative descriptor, executes a processing operation corresponding to the first operator identifier to obtain intermediate data, and directly transmits the intermediate data to a storage area of the SmartNIC through a PCIe P2P data path; and the SmartNIC executes a processing operation corresponding to the second operator identifier on the intermediate data to obtain target data, and transmits the target data to the computing device according to the target address information in the collaborative descriptor.

The invention provides a subject intelligent evaluation informatization system based on medical record home page data. Comprising a data integration and treatment module, a data standardization processing module, a core index intelligent calculation engine, a subject competitiveness multi-dimensional evaluation model, a DRG / DIP analysis application module, a medical quality and safety monitoring module, an intelligent report generation and visualization platform and a system management and configuration center. Through full-process automatic and intelligent analysis, multi-source data can be automatically integrated and strict quality control and standardization processing can be carried out, the accuracy and consistency of evaluation data are guaranteed from the source, complex analysis from basic index operation to comprehensive competitive power score ranking can be rapidly and accurately completed, and the method is suitable for large-scale popularization and application. And the result is visually presented through a visual platform, and even a structured analysis report is automatically generated, so that managers are liberated from tedious data processing and can concentrate on decision analysis, and the management efficiency and the response speed are remarkably improved.

The invention provides a fault detection method and device for an intelligent computing platform, electronic equipment and a medium, and the method comprises the steps: obtaining the heterogeneous data of at least one of an infrastructure layer, a platform service layer and a business system layer in the intelligent computing platform, and carrying out the fusion processing of at least one piece of heterogeneous data, and obtaining a target data flow; decomposing at least one of a fault analysis task, a fault positioning task and a fault recovery task of the intelligent computing platform according to the target data stream to obtain at least one subtask; determining a diagnosis context according to the target data flow, and determining at least one model service through the diagnosis context; and obtaining a fault detection result or a fault recovery result of the intelligent computing platform according to the diagnosis context, the at least one subtask and the at least one model service.

The invention, which belongs to the technical field of data processing, relates to an intelligent convenient gravel aggregate transaction system comprising a user interaction layer used for receiving a user instruction and displaying information; the cloud intelligent central layer is in communication connection with the user interaction layer and is used for performing intelligent calculation, data collaboration and safety control; and the full-link system docking layer is in communication connection with the cloud intelligent center layer and is used for realizing data intercommunication with an external service system. Through technology fusion of AI, the block chain and the Internet of Things and whole-process collaborative design, information islands are eliminated, automatic closed loop of the whole transaction process and whole quality traceability are guaranteed, and the transaction risk is reduced.

The invention relates to the technical field of optimal configuration control, in particular to a power grid planning-oriented artificial intelligent computing resource optimal configuration control method, which comprises the following steps of: establishing an association mechanism of a power grid planning scene and load characteristics; based on a power grid planning scene and load characteristics, constructing an accurate calculation power demand measurement and calculation model; intelligent computing resource collaborative scheduling is realized from three dimensions of a hardware layer, a software layer and a task layer; according to the method, through scene and load mapping, the matching degree of resource configuration and a power grid planning scene is improved, and the problem that universal configuration is not applicable is solved; the measurement and calculation error of a domain-specific computing power formula is reduced, and resource excess or insufficiency is avoided; after multi-dimensional optimization, the intelligent computing resource utilization rate is improved, the model training / reasoning delay is reduced, and the AI application cost of power grid planning is remarkably reduced.

The invention discloses a manufacturing industry production data real-time management system and method based on Internet of Things edge computing. Comprising a data acquisition and preprocessing module, a lightweight agent module, an edge computing module, a network transmission module, a platform management module, an application service module, a security protection module, a self-adaptive resource configuration module and a monitoring and diagnosis module. According to the invention, nearby acquisition, preliminary cleaning and real-time analysis of multi-source heterogeneous data from a sensor, a controller and industrial equipment are realized, the cloud burden is effectively reduced, and the overall processing efficiency is improved. A real-time anomaly detection mechanism based on a sliding window and machine learning is constructed, parameter deviation, equipment faults or process anomalies in the production process can be recognized within millisecond-level time, and early warning or control feedback is triggered.

The invention discloses a cloud edge end collaborative computing power scheduling system supporting task dynamic migration, which relates to the technical field of intelligent computing power scheduling and comprises a path modeling module, a conflict identification module, a time sequence checking module, a rehearsal switching module and an interference suppression module, and constructing a routing convergence time sequence profile and a link playback spectrum, modeling an overlapping window of the old path and the new path and a life cycle of a control signal, and generating a path evolution baseline for subsequent conflict identification. By means of path modeling, fingerprint identification, causal check, shadow rehearsal, interference suppression and the like, time sequence consistency of a control instruction and a data stream, continuous path switching and real-time extinguishing of link interference are realized, and the stability and reliability of cloud edge collaborative scheduling are improved.

The invention provides a supply chain bill ecological interconnection system based on an open API (Application Program Interface) and a smart contract, which relates to the field of electric digital data processing and comprises a self-adaptive open interface collaboration module, a cross-chain smart contract arrangement module, a multi-source credit intelligent calculation module and an end-to-end privacy security protection module, the self-adaptive open interface collaboration module is responsible for realizing intelligent access and authority management of a heterogeneous system, the cross-chain intelligent contract arrangement module is responsible for realizing unified contract management and automatic execution in a multi-block chain environment, and the multi-source credit intelligent calculation module is responsible for fusing multi-party data to construct a supply chain credit evaluation system. The end-to-end privacy security protection module is responsible for providing data encryption, privacy calculation and identity verification of the whole process; according to the system, the problems of data islands, unsmooth credit transmission, unbalanced privacy protection and data utility and the like in a supply chain bill system can be solved, and a complete technical solution is provided for digital ecology of supply chain finance.

The invention discloses an intelligent computing center resource management and control method, system and device based on computing and power collaboration. The method comprises the steps of collecting computing power resource state data of an intelligent computing center, time-of-use electricity price data of an external power grid, state data of a micro-grid and state data of an energy storage system in real time; fusing the computing power resource state data, the time-of-use electricity price data, the microgrid state data and the energy storage system state data to obtain fused data; predicting the fusion data by using a pre-trained large model, and generating a computing power resource demand prediction curve and a microgrid power generation capacity prediction curve in a future set time window; then, by taking the minimization of the total power consumption cost as an optimization target, generating a power calculation cooperative scheduling strategy; and according to the calculation and power cooperative scheduling strategy, performing calculation power resource distribution, microgrid output adjustment and energy storage system charging and discharging control. The method can effectively break the separation barrier of computing power and electric power, and realizes deep cooperation of computing power scheduling and energy management.

An artificial intelligence (AI) calculation circuit is provided. The AI calculation circuit can support various integer and floating-point calculations through the adjustment of circuit configuration. Integer multiplication and floating-point mantissa multiplication share the multiplication unit, integer comparison and floating-point comparison share the same comparison unit, integer addition and floating-point addition share the same addition unit.

The invention belongs to the field of bridge main cable structure health monitoring, and provides a GNSS-based long and large bridge main cable self-driven high-precision sensing method and device, and the method comprises the steps: collecting GNSS original displacement data, and carrying out the timing preprocessing through a median absolute deviation (MAD) algorithm; using a CNN-LSTM hybrid model to filter out non-structural noise such as a multipath effect; a reference response model is constructed by adopting an adversarial space-time Transform network, and abnormal signals are accurately separated; the sampling and communication frequency is adaptively adjusted according to the environment state; online optimization of the model is realized by an abnormal driving incremental learning mechanism; the device comprises a data acquisition and preprocessing module, an edge intelligent calculation module, an energy-saving management module and an incremental learning module. Millimeter-level high-precision sensing, low-power-consumption self-driven operation and long-term evolution capability are realized, and the intelligent level of bridge safety monitoring is remarkably improved.

The invention belongs to the technical field of ocean engineering. According to the seabed data intelligent calculation center structure and the working method, a plurality of stand columns and a tubular structure are connected to form a frame used for bearing a tank body, a supporting seat is fixed in the frame, and the tank body is arranged on the supporting seat; grating plates are arranged at the two ends and the top of the frame, a plurality of suction foundations are fixed to the bottom of the frame, and data intelligent calculation equipment is arranged in the tank body; a temperature and humidity real-time monitoring sensor and a pressure balance adjusting module are integrated in the tank body, the temperature and humidity real-time monitoring sensor communicates with a cooling system outside the tank body, the heat dissipation efficiency in the tank body is adjusted in a linkage mode according to real-time temperature and humidity data, and the pressure balance adjusting module is used for automatically compensating air pressure in the cabin according to water pressure differences of different depths of the seabed; according to the invention, the structural stability of the seabed is ensured, natural cooling is carried out through low-temperature seawater, the problem of high heat dissipation energy consumption of a traditional data center is solved, and a stable working temperature environment is provided for data intelligent calculation equipment.

The invention relates to the technical field of data processing, and discloses an intelligent computing center energy consumption optimization method and system based on the Internet of Things. The method comprises the steps of collecting storage medium energy consumption parameters and data block access behavior parameters to generate an original data set, calculating a weighted access characteristic value and a cold and hot state transition probability, screening candidate data blocks and calculating an energy consumption benefit evaluation index, matching a renewable energy source supply state to construct a migration task sequence, and obtaining a new energy consumption benefit evaluation index. And executing migration operation and dynamically adjusting the prediction parameters. The problems that in the prior art, data cold and hot state conversion cannot be intelligently predicted, energy consumption benefit quantitative evaluation is lacked, cooperation of data migration and renewable energy source supply is insufficient, and prediction model parameters cannot be adaptively adjusted are solved.

The invention belongs to the technical field of computing power network scheduling, and particularly relates to an intelligent computing power network scheduling system based on digital simulation. Comprising an agent knowledge support module, a multi-agent self-organizing interaction module, a trend evolution simulation module and a macroscopic regulation and control evaluation optimization module. The agent knowledge support module provides support for dynamic evolution of the agent by constructing a multi-level and structured knowledge fusion system; the multi-agent self-organizing interaction module realizes distributed collaborative decision based on a reinforcement learning framework and a game equilibrium theory; the trend evolution simulation module is responsible for performing accurate prediction and risk assessment on the future state of the computing power network; the macroscopic regulation and control evaluation optimization module automatically searches an optimal policy combination based on a multi-objective optimization algorithm and provides data support for a decision maker; the system modules cooperate with each other to form a data-driven closed-loop feedback updating mechanism. The resource waste is reduced, and the scheduling efficiency is improved.

A system for cleaning, zeroing, and preventing air pollution in indoor space includes several gas detection devices, a central control monitoring device, and several gas exchange and filtration devices. The gas detection devices detect and output indoor and outdoor air pollution data. The central control monitoring device receives the outdoor and indoor air pollution data to perform an intelligent computation and comparison, locate an air pollution location in the indoor space, and transmit a control command intelligently. Each of the gas exchange and filtration devices is combined with a smart switch for the control command to control the gas exchange and filtration devices to be enabled, so that the air pollution is filtered and exchanged by the gas exchange and filtration devices, thereby allowing the indoor air pollution data detected by the gas detection devices in the indoor space to be a safety detection value.

The invention provides a unified computing power management and control platform and method.The computing power management and control platform is integrated with a plurality of intelligent computing centers in different position areas through a unified northbound interface, a resource management module abstracts physical resources of the intelligent computing centers into virtual resource data, and a unified resource pool is constructed; the data acquisition module receives operation monitoring data of each intelligent calculation center and preprocesses the operation monitoring data; the resource sensing module is used for constructing a dynamic resource portrait for the uniform resource pool based on the preprocessed operation monitoring data; the task sensing module performs resource constraint and service quality demand analysis on a calculation task submitted by a user, and constructs a task portrait; the resource scheduling module forms a resource scheduling decision result according to the dynamic resource portrait and the task portrait; and the resource arrangement module performs resource scheduling and calculation task allocation according to the resource scheduling decision result. The overall utilization rate of the heterogeneous computing power cluster is remarkably improved, and the island effect of a traditional resource pool is broken through.

The embodiment of the invention discloses a task execution method and system. The method comprises the following steps: in response to a to-be-executed artificial intelligence task, identifying task features of the artificial intelligence task, and obtaining a first resource portrait of a general computing domain and a second resource portrait of an intelligent computing domain in a hybrid cloud environment; determining a first matching degree between the task feature and the first resource portrait, and determining a second matching degree between the task feature and the second resource portrait; and according to a numerical relationship between the first matching degree and the second matching degree, determining a target computing domain from the general computing domain and the intelligent computing domain, and allocating the artificial intelligence task to the target computing domain, so that the target computing domain is utilized to execute the artificial intelligence task. According to the technical scheme provided by the embodiment of the invention, the optimal balance between the cost and the performance can be realized.

The invention discloses an intelligent computing center-oriented multi-type hardware equipment energy efficiency evaluation system, which realizes dynamic and accurate evaluation of individual and overall collaborative energy efficiency by collecting full-dimensional data of multi-type hardware in an intelligent computing center, and further quantifies the influence of resource interaction on the energy efficiency by establishing a cross-equipment collaborative relation model, thereby improving the energy efficiency evaluation efficiency. A targeted optimization strategy is generated, and a closed-loop management mechanism integrating energy efficiency evaluation, intelligent decision, instruction execution and data traceability is finally formed, so that continuous optimization and improvement of the energy efficiency of the intelligent calculation center are comprehensively driven; the system comprises a multi-type hardware data acquisition module, a dynamic load feature extraction module, a multi-dimensional energy efficiency evaluation module, a cross-device collaborative modeling module, a dynamic optimization decision module, a visual interaction module and a data storage and traceability module.

The invention provides a unified computing power intelligent scheduling system and method. A demand quantification module converts computing demand information of a computing task into a demand feature vector; the factor quantification module quantifies decision factors of resource scheduling; the weight adjustment module dynamically adjusts the weight coefficient of each decision factor; the resource decision-making module is used for solving a multi-target optimization scheduling decision-making model based on the demand feature vector and the dynamic resource portrait to obtain a resource scheduling decision-making result of the calculation task, and the multi-target optimization model is constructed based on each quantized decision-making factor and a weight coefficient of the decision-making factor. Unified coordinated scheduling of multiple intelligent computing centers is achieved, the problem that in the prior art, scheduling optimization is only based on a single-dimension (only performance and only utilization rate) single intelligent computing center is overcome, balance among multiple targets is achieved, the overall efficiency of the system is maximized, the system adapts to complex heterogeneous environments and dynamic changes, and the robustness and fault-tolerant capability of the system are improved.

The invention relates to a high-performance spaceborne intelligent calculation terminal and a starting method thereof. The terminal comprises a routing unit, a storage unit and an intelligent calculation unit, each of the routing unit and the intelligent calculation unit comprises a hardware physical layer, a hardware driving layer, an embedded operating system, a resource scheduling management layer and an application service layer; the application service layer is used for configuring different application service APPs according to spaceflight application scenes and function requirements, and the application service APPs comprise service plug-ins corresponding to the resource scheduling management layer called by an embedded operating system and universal task distribution and resource management modules in the operating system according to the different application service APPs related to intelligent AI model intelligent calculation. In combination with corresponding hardware drive, computing power and hardware resources required by the intelligent AI model are flexibly configured for multi-user application software; the routing unit is used for providing interface communication with the outside, and realizing optical fiber data receiving and transmitting health management, Ethernet bus communication, RS422 bus communication, data routing management, task scheduling management, remote control and telemetering, centralized management of a system program or an application program or an FPGA configuration program, and a power supply control function; the intelligent calculation unit dynamically loads the on-satellite intelligent AI model and parameters thereof and is used for completing operation of the on-satellite intelligent AI model; and the storage unit is used for storing on-satellite load original data, intermediate data processed by the on-satellite intelligent AI model and processed data.

A system and method for intelligent computerized task scheduling and execution, including: optimizing a time off rule including a quota of time off units for a time period—by changing the quota of time off units based on calculating a time off utilization indicator; updating a computerized task schedule based on the optimized time off rule; and executing tasks based on the updated schedule. In some embodiments, time off optimization may include identifying, by a machine learning model (such as, e.g., a generative artificial intelligence or large language model), rules matching a given time off rule, and deleting / merging rules based on similar rule names or activity codes. The machine learning model may generate rule names for merged rules. Optimized time off rules may be used to accept or reject time off requests transmitted and / or received, e.g. over a data or communication network.

The present disclosure relates to a data processing method, a processor, a chip, a graphics card and an electronic device, and relates to the technical field of intelligent computing. The method comprises the following steps: in response to M tensor computing engines in a tensor computing engine cluster completing reading of a partial first matrix from a local memory or a register of a computing unit to which the M tensor computing engines belong and reading of a second matrix from a tensor memory, the M tensor computing engines respectively perform matrix operation according to the partial first matrix and the second matrix to obtain corresponding operation results of the M tensor computing engines; and the M tensor computing engines respectively write the corresponding operation results into the registers of the M tensor computing engines. The embodiment of the present disclosure can increase computing power without increasing the size and bandwidth of the local memory and the register inside the computing unit.

The invention provides a computing power unified charging system, which is arranged on a unified computing power management and control platform, the unified computing power management and control platform is integrated with a plurality of intelligent computing centers at different geographic positions through a unified northbound interface, and a policy configuration module performs charging policy configuration; the usage analysis module performs resource usage analysis on the operation monitoring data acquired from each intelligent calculation center to obtain resource usage data of the user, wherein the resource usage data is composed of usage of each single resource; the unit price calculation module adjusts the basic unit price of each single resource based on the multiple adjustment factors corresponding to the single resource to obtain a billing unit price; and the resource charging module is used for calculating the total cost of the user by adopting a charging strategy matched with the user based on the resource consumption data of the user and the charging unit price of each single resource in the resource consumption data. And efficient integration and unified and flexible charging of dispersed intelligent computing resources are realized.

The application discloses a kind of face depth forgery detection methods based on multi-modal fusion, belong to artificial intelligence, computer vision and multimedia forensics field.For the generalization deficiency of existing detection method, audio and picture time sequence inconsistency problem, end-to-end detection scheme is proposed.The method fuses the time-aligned audio spectrum and continuous video frame mosaic picture in advance, forms multi-modal unified input to solidify audio and picture time sequence relationship.Global branch uses contrast learning pre-training visual Transformer, introduces modal bias and cross-modal attention bias to enhance feature distinction interaction;Local branch fuses local and global features through multi-scale region perception network, realizes adaptive weighted fusion by combining position coding, vector-level channel attention and scale attention, and is guided by region perception auxiliary loss to select discriminative features.The method provides an effective tool for content review, traceability forensics and identity authentication security.

The invention provides a multi-dimensional charging method and device for computing power resources and electronic equipment, and the method comprises the steps: converting physical measurement indexes of all resource types in an original resource ticket, and obtaining a standard resource ticket; constructing a rate configuration model based on the basic unit price and the adjustment coefficient corresponding to each resource type; the adjustment coefficient is used for reflecting the resource commercial value difference of each resource type; and finally, inputting the standard resource ticket into the rate configuration model to obtain a cost result. According to the method provided by the invention, the original resource ticket of each resource type is acquired and converted into the standard resource ticket, so that unified measurement of heterogeneous resource data is realized; by constructing the rate configuration model containing the resource performance coefficient and comprehensively considering the influence of the resource type and the performance difference on the commercial value, accurate metering and charging of various intelligent computing resources are realized, the benefits of service providers and users are effectively balanced, and the standardization and high efficiency level of intelligent computing power business charging management is improved.

The invention provides a method and a device for automatically detecting automobile drawing size marking by an intelligent computing cloud platform through computing power, and relates to the technical field of intelligent computing cloud platforms and computing power infrastructure, the method comprises the following steps: constructing a sample data set based on at least one of a first standard parameter, a second standard parameter and a third standard parameter, dividing the sample data set into a training set and a verification set; the intelligent computing cloud platform performs N times of iterative training on the initial model based on the training set by using computing power to obtain a final model; and deploying the final model on the intelligent computing cloud platform so as to carry out size marking detection on the received automobile drawings in batches. The intelligent computing cloud platform performs automatic detection of the automobile drawing size marks through computing power, can accurately position the marks and identify the size mark abnormity in the automobile drawing, reduces the understanding difference in the manual inspection process of the automobile drawing size marks, and improves the detection efficiency and accuracy.