111results about How to "Strong computing power" patented technology

The invention provides an intelligent industrial quality inspection method and system based on AI visual identification, and relates to the technical field of visual identification, and the method comprises the steps: inputting an industrial product image collected at an edge end into a pre-trained defect region detection model, and obtaining a defect region detection result; extracting defect area features according to a defect area detection result, carrying out dimension reduction processing to obtain compressed defect area features, and deciding to unload a defect classification task to a cloud end or directly carry out defect classification at the edge end according to an adaptive task unloading strategy in combination with working data of the edge equipment; if the defect classification task is determined to be unloaded to the cloud, transmitting the compressed defect area features to the cloud through a network, performing defect type identification on the cloud by adopting a pre-trained defect classification model, outputting a classification result of the defect area, returning the classification result to an edge end, and combining a defect area detection result and the classification result of the defect area to obtain the defect classification task. And generating a defect visualization effect picture and a quality inspection report.

The invention discloses an industrial visual detection system based on multi-modal fusion, which belongs to the technical field of industrial data acquisition, processing and processing, and comprises a multi-modal data acquisition and calibration module, a scene image capture module, a visual information synthesis module, an algorithm optimization and training module and a multi-modal image output module, the multiple modules are integrated into the industrial visual detection system for multi-modal data acquisition, processing and fusion, high-quality original data are provided for subsequent image analysis, images of industrial scenes can be automatically captured under different illumination conditions, an original image database containing various illumination and scene conditions is constructed, and the visual detection system has the advantages of being simple in structure, convenient to operate and high in practicability. The system can operate stably, high-precision analysis and processing of the image data set in the industrial scene are realized, the corresponding fusion image is generated, the industrial visual inspection report is generated based on the fusion image, and the adaptability and reliability of the industrial visual inspection system are improved.

The invention relates to the technical field of metal materials, and provides a medium-and-high-entropy aluminum alloy performance prediction method, system and method and a medium, and the method comprises the following steps: obtaining a characteristic data set for medium-and-high-entropy aluminum alloy performance; constructing an MLP prediction model, and obtaining an optimal hyper-parameter combination of the MLP prediction model through ESOA; training the MLP prediction model according to the feature data set and the optimal hyper-parameter combination; and predicting the performance of the medium-high entropy aluminum alloy according to the trained MLP prediction model. According to the method, the advantages of the ESOA and the MLP prediction model are combined, the accuracy of performance prediction of the medium-high entropy aluminum alloy can be effectively improved, the time and experiment cost can be reduced, the prediction cost and the calculation amount can be reduced through the powerful calculation capacity of the MLP prediction model, the prediction precision and the generalization capacity of the model are enhanced, and the method is suitable for popularization and application. The method is suitable for performance prediction of the medium-high-entropy aluminum alloy material.

The invention relates to the technical field of computing power infrastructure, and provides a document analysis method and device based on computing power resources of an intelligent computing center, and the method comprises the following steps: S1, obtaining a to-be-processed document in a first format, and converting the document in the first format into a document in a second format, S2, inputting the document in the second format into a layout recognition model, obtaining a layout recognition result of the document, and inputting the layout recognition result into a content recognition model to obtain content information of the document; and S3, combining the layout recognition result and the content information to obtain an analysis result of the document. Therefore, the layout recognition model and the content recognition model which are operated based on the computing power resources of the intelligent computing center are used for analyzing the document, the document analysis speed can be remarkably increased, and the text semantic understanding accuracy and the text information extraction accuracy are greatly improved.

The invention discloses a quantum adversarial attack method based on momentum, which refers to an object momentum mechanism, and designs a quantum adversarial attack method to efficiently generate adversarial samples. In order to solve the problem that an existing quantum countermeasure attack method excessively fits a specific model to cause a poor local optimal solution, the invention provides a novel quantum countermeasure attack method based on momentum (Q-MIFGSM). According to the method, a loss function of a quantum neural network classifier and momentum accumulation gradient information of a sample are defined, a momentum iteration updating mode is adopted, momentum accumulation is carried out on a loss function gradient value, the optimization direction is dynamically adjusted in combination with current and historical gradients, and a local optimal solution is effectively jumped out; the loss function value of the quantum classifier is increased to cause classification errors of the classifier, so that attack resistance is realized, and an adversarial sample is efficiently generated. Simulation experiment results show that the Q-MIFGSM algorithm can effectively execute adversarial attacks on the quantum neural network classifier and generate adversarial samples close to original data. A new strategy is provided for quantum attack resistance, and the method has potential application value in the aspects of quantum neural network security evaluation and defense strategies.

The invention discloses a matrix calculation unit optimization-based sparse LU decomposition solving method and a related device, and relates to the technical field of numerical algorithm optimization, and the method comprises the steps: calling an improved PRCM algorithm to carry out the reordering of a sparse matrix, improving the local density of data, obtaining a sparse matrix after bandwidth compression, and obtaining a sparse matrix; then converting the sparse matrix into a regular and dense Tile block structure, storing the Tile block structure in a pre-allocated continuous memory space, performing numerical decomposition on the sparse matrix after bandwidth compression by adopting a right view-based LU decomposition algorithm, and accelerating by using a matrix calculation unit in the numerical decomposition process to obtain an LU matrix; therefore, efficient solving of the to-be-solved linear equation can be completed. According to the scheme, the spatial locality of the sparse matrix is fully utilized, the strong computing power of the matrix computing unit is exerted, the computing efficiency is improved while the computing amount is reduced, and the time of sparse LU decomposition in symbol decomposition and numerical decomposition stages is remarkably shortened.

The invention provides a cloud computing-based narrow-linewidth 1200 nm semiconductor laser data processing system, and relates to the technical field of data processing, and the system comprises a laser transmitting module which is used for transmitting a narrow-linewidth 1200 nm laser beam by using a semiconductor laser; the data acquisition module is used for being connected with the laser emission module, acquiring state parameters, including power, wavelength, line width and working environment data, of the laser beam in real time, and preprocessing and digitally converting signals output by the sensor; and the cloud processing platform module is used for receiving and processing the data of the data acquisition module, realizing real-time analysis, anomaly detection, trend prediction and remote monitoring of the data, and obtaining an analysis result of a cloud processing platform. According to the invention, remote intelligent monitoring and automatic optimization adjustment of the laser are realized, and the data processing efficiency and the performance stability of the laser are improved.

The invention relates to an improved FPGA (Field Programmable Gate Array)-based storage system platform and an operation method thereof. The storage system platform comprises an FPGA (Field Programmable Gate Array) master control module which is mounted with a DDR4 (Double Data Rate 4) memory module and serves as a control core of the storage system platform to control high-speed reading of data of external equipment; the communication module is used for reading data of external equipment through an Aurora communication interface and transmitting the data outwards through a 10-gigabit optical port; the NVMe storage module interacts with the FPGA main control module through a hard core PCIe bus and stores external data; the health management module monitors the operation state of the storage system platform; the clock module is used for providing clock management for the work of the storage system platform; the power supply module is used for providing power management for the work of the storage system platform; the FPGA main control module is provided with a file management system used for file management and assisting the FPGA main control module in data management.

The invention relates to a commanding height analysis method and system based on double spatial indexing, and the method employs a double index construction method through an index construction link, i.e., constructing double indexes of a tile quadtree and a triangular patch K-D tree, thereby achieving the quick positioning of a triangular patch intersecting with a ray in a live-action three-dimensional model. By means of the powerful operational capability of a computer, potential hidden danger commanding height points along the security and protection route can be accurately found, the commanding height point calculation efficiency is greatly improved, the practical feasibility of completing large-scale commanding height point calculation based on a live-action three-dimensional model is enhanced, and security and protection work is promoted to be developed more scientifically and efficiently.

The invention provides a measurement method and device based on an unmanned aerial vehicle and edge computing equipment. The measurement method comprises the steps of controlling the unmanned aerial vehicle to cruise a distribution wire according to a preset route; the unmanned aerial vehicle is controlled to carry out image acquisition on the power distribution wire through a camera in the cruising process to obtain a line image; the unmanned aerial vehicle is controlled to obtain spatial position coordinates of at least three key points on the power distribution wire and spatial position coordinates of base poles located at the two ends of the at least three key points in the cruising process; and preprocessing the line image to obtain a processed image. The technical problem solved by the invention is that sag measurement of a power distribution wire generally depends on a traditional manual measurement method in the power industry. Although the method meets the basic measurement requirement to a certain extent, the inherent limitation of the method is increasingly prominent, and the method becomes an important bottleneck restricting the improvement of the power operation and maintenance efficiency and precision.

The invention discloses a large model reasoning method and system, computer equipment and a storage medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: segmenting a pre-trained target large model, and obtaining an input layer, a middle layer and an output layer corresponding to the target large model; deploying the input layer and the output layer at a client, and deploying the intermediate layer at a server; receiving a Prompt input by a target user at the client, and encoding the Prompt through the input layer to obtain a corresponding numerical vector; uploading the numerical value vector to the server, and reasoning the numerical value vector through the intermediate layer to obtain a corresponding result vector; and returning the result vector to the client, and decoding the result vector through the output layer to obtain a corresponding result text. By means of the method, privacy protection in the large model reasoning stage can be achieved.

The invention provides a navigation computer for a small inertial navigation system, the size of the navigation computer is only 55mm * 86mm * 16mm, a navigation resolving module, an I / F conversion module and a power supply module are integrated, and the power consumption is less than 3W; the navigation resolving module takes a ZYNQ7000 all-programmable SoC as a core processor, and has the functions of inertial device original data acquisition, navigation information preprocessing, navigation resolving and the like; the I / F conversion module can directly convert a current signal of the accelerometer into a pulse signal, and the pulse signal is accessed to the navigation resolving module for navigation resolving; the power supply module can output + 5V (8W),-5V (3W), + 15V (9W) and-15V (3W) isolated voltage-stabilized power supplies to the outside, and can supply power to other loads of the inertial navigation equipment. The method is suitable for laser inertial navigation equipment with small volume, low power consumption and high reliability.

The invention relates to a block chain consensus method for the Internet of Things, and the method specifically comprises the steps: carrying out the statistics of the data collection frequencies of all data collection nodes in the Internet of Things, and determining a consensus target value according to the number of consensus nodes and the frequencies; the method comprises the following steps: acquiring network flow and a source of the network flow of a consensus node from last block generation to the current moment, determining the participation degree of the consensus node in the Internet of Things by utilizing the network flow, the source of the network flow and the type of the consensus node, taking a node of the source of the network flow as a verification node, and verifying the participation degree by the verification node, if the verification is not passed, the reputation value of the consensus node is adjusted based on the data volume sent to the consensus node by the verification node; and the consensus node recalculates the target value by using the reputation value and the participation degree to obtain the target value of the consensus node when the block is generated, generates the block based on the target value of the consensus node and adds the block into the block chain. The problem that resources are limited when the block chain is applied in the Internet of Things is effectively solved.

The invention relates to an FPGA (Field Programmable Gate Array)-based surfacing workpiece surface defect detection method and device. The method comprises the following steps: acquiring a magnetic induction intensity data matrix; performing phase-locked amplification on the magnetic induction intensity data matrix to obtain an original feature data matrix; performing convolution filtering processing on the original feature data matrix to obtain a smooth feature data matrix; performing binarization processing on the smooth feature data matrix to obtain a binarized data matrix; whether magnetic field distortion exists on the surface of the surfacing workpiece or not is determined according to the binary data matrix; and when the magnetic field distortion exists on the surface of the surfacing workpiece, judging whether the magnetic field distortion is caused by surface defects of the workpiece or not, and outputting a detection report according to a judgment result. Defect detection is realized by utilizing hardware, the advantage of high operational capability of the FPGA is fully utilized, and the efficiency of defect detection is remarkably improved.

The invention belongs to the technical field of DC-AC converters, and relates to a single-stage current feed type bidirectional DC-AC converter and a fixed frequency control method thereof.The single-stage current feed type bidirectional DC-AC converter comprises a DC side composed of two Boost half bridges with inductors, an AC side composed of half bridges and an AC side filter, a transformer and a control device; the control device is used for controlling the direct current side and the alternating current side to describe the power transmission characteristics of the circuit by transconductance Gm, and the current control of the alternating current side is realized by a feedforward method; fixed-frequency linear phase shift control can be carried out in an ultra-wide DC side voltage range, bidirectional real-time power flow control can be realized in a relatively wide DC side voltage range, and a complex control loop design theory is not needed; therefore, the cost and the power consumption of the controller are effectively reduced, and a higher power density upper limit and a higher efficiency upper limit are achieved.

The invention relates to the technical field of navigation signal detection, and discloses a small portable navigation signal comprehensive detection system and method.The system comprises a building block type mobile antenna support, backpack type comprehensive detection equipment and a navigation equipment site evaluation system, the system is composed of a backpack type comprehensive detection host and an individual soldier handheld terminal, and rapid opening and comprehensive detection of navigation signals are achieved. The backpack comprehensive detection host integrates a battery, a power supply module, a high-performance signal processing board, a radio frequency channel and a medium-wave active antenna, adopts a high-performance battery and power supply management scheme and an advanced AC7Z020 core board for signal processing, and ensures high-precision and high-sensitivity detection. The system is further integrated with an electronic compass module to achieve a direction finding function, real-time data feedback and evaluation are achieved through Bluetooth communication between the individual soldier handheld terminal and the host, the detection convenience and reliability are greatly improved, and the system is particularly suitable for scenes such as field environments, emergency rescue and military actions.

The invention relates to a novel industrial cloudization controller undisturbed upgrading method and system and a medium, and the method comprises the steps: creating a cloudization controller; the uploaded cloud controller software mirror image is received and uploaded to the private cloud platform, and mirror image preheating is carried out after verification; in response to the upgrade request, identifying and selecting a standby controller to upgrade by referring to the preheated mirror image, and controlling the standby controller to obtain and update a configuration preheating memory in the corresponding main controller; and switching the upgraded standby controller into the main controller for pilot run, switching the main controller into the standby controller, and continuing to upgrade the original main controller or rolling the upgraded standby controller back to the version before upgrading according to the running state of the current control service in the pilot run. According to the method, the hardware controller is clouded, active and standby undisturbed upgrading and flexible resource management are realized, the system flexibility is improved through an intelligent process, the maintenance cost is reduced, and powerful support is provided for efficient operation and maintenance of the industrial internet.

The invention provides a structured data processing method and device based on a cloud mode, a medium and equipment. The method comprises the following steps: receiving structured data sent by equipment, and storing the structured data in a database; receiving a data query request sent by a user; analyzing the data query request to obtain query parameters; selecting an SQL statement template corresponding to the data query request from a template library; wherein SQL (Structured Query Language) statement templates in a plurality of business scenes are stored in the template library; the SQL statement template is filled with the query parameters, and SQL statements are obtained; and executing the SQL statement to query the required structured data from the database, and feeding back a query result to the user. According to the method, the distributed computing capability of cloud computing and the flexibility and the real-time performance of the end equipment are combined, the processing efficiency of the structured data can be improved, the cost is reduced, and more convenient and safer data processing services are provided for users.

The invention discloses a hyper-arithmetic scheduling server and a resource management method, and relates to the technical field of communication, the method comprises the following steps: 1, scheduling server data acquisition: acquiring server task requirements and resources acquired by a system; 2, preprocessing and classifying the collected server resource information; 3, performing classified management on effective resources; 4, constructing and updating an original scheduling strategy according to task requirements of workers; 5, monitoring server resources in real time; and 6, protecting the resource information of the scheduling server. According to whether the monitored resources are in the safe state or not, an alarm mechanism is established, related data are recorded, an external environment manager is warned of dangerous situations, the situation of resource leakage is blocked in time, losses caused to people are avoided, and the safety of the computing power scheduling server system is effectively guaranteed.

The invention belongs to the technical field of semantic communication, and particularly relates to a hybrid semantic communication system based on a pulse neural network. The method comprises the following steps: performing semantic extraction on a server side by using an ANN model; then, the semantic information is converted into a data format suitable for SNN model processing after being subjected to time-space dimension semantic discretization; the received semantic information is processed at the edge end by utilizing an SNN model which is easy to deploy and low in power consumption, so that efficient and high-precision semantic understanding is realized. And different frequency characteristic optimization distribution modes of the intelligent distribution sub-module are matched with corresponding pulse neurons to trigger calculation only when a pulse event is detected, and the threshold condition of multiple signal triggering of a single pulse neuron can be set, so that as many as three levels of triggering type power consumption optimization can be realized. Under the condition that the FPGA is combined and used, high-efficiency performance is kept when large-scale data are processed. The method is particularly suitable for low-power-consumption and efficient communication application scenes.

The invention discloses a short-circuit protection device of a GaN power switch device. The short-circuit protection device comprises a GaN power switch; the current detection resistor is connected between the source electrode of the GaN power switch and the ground and is used for detecting the current in the circuit; the input end of the comparator receives the amplified current signal and the voltage signal subjected to voltage division processing respectively, and the comparator is used for comparing the two signals and outputting a comparison result; the driving circuit is used for receiving a control signal from the microcontroller so as to control the grid voltage of the GaN power switch; the microcontroller is used for monitoring the output of the comparator and controlling the driving circuit according to a monitoring result, so that the GaN power switch is controlled; the microcontroller further reads the output of the comparator through the ADC so as to monitor the current and voltage states of the circuit in real time, and corresponding protection measures are taken when abnormity is detected. According to the invention, effective management and protection of the GaN power switch are realized, and powerful support is provided for safe operation of a power electronic system.

The invention discloses a battery voltage difference fault diagnosis method, device and equipment and a readable storage medium, and the method comprises the steps: a cloud diagnosis platform obtains vehicle battery data, and the vehicle battery data comprises current static voltage data; calculating a difference value between the maximum charge state and the minimum charge state of each single battery in the current static voltage data to obtain a current charge state difference value; calculating the growth rate of the current charge state difference value relative to the charge state difference value before the first preset duration; and performing battery voltage difference fault diagnosis according to the current charge state difference value and the growth rate. Compared with the prior art, the method has the advantages that the cloud diagnosis platform has higher data storage and calculation capability compared with a battery management system, the cloud diagnosis platform calculates the current charge state difference value and the growth rate according to the obtained latest vehicle battery data, and further performs battery voltage difference fault diagnosis based on the current charge state difference value and the growth rate; therefore, timely and effective battery voltage difference fault diagnosis management is realized.

The present invention relates to a driving state detection and early warning system and method based on multimodal technology, so as to solve the technical problem that the driver is prone to red vision or black vision during high-speed driving, which leads to serious driving accidents. The system includes a helmet unit and a remote control module; the helmet unit includes a shell, an eye tracking detection module, a cerebral blood oxygen detection module, a signal processing module, a voice interaction module, a power supply module and a reference coordinate construction module. The method includes: 1. Wearing a helmet unit, constructing a reference coordinate system of the driving area, and dividing at least one high-heat area; 2. Starting each module of the helmet unit, real-time detection of the driver's eye capillary texture characteristics, attention mechanism distribution characteristics and cerebral blood oxygen real-time data characteristics, and issuing a warning signal according to the detection results. The present invention conducts a real-time safety assessment of the driver's state and issues a warning through multi-channel, multi-dimensional and multi-modal analysis.

The application relates to a CAD drawing sharing and discussing system, which comprises a first terminal, one or more second terminals, a first CAD drawing sharing module and a first instant messaging module in the first terminal, a second CAD drawing sharing module and a second instant messaging module in each of the one or more second terminals, the first CAD drawing sharing module and the second CAD drawing sharing module transmitting and sharing CAD drawings with each other, and the first instant messaging module and the second instant messaging module transmitting instant messaging information with each other. In the application, each terminal can look at the CAD drawing while instant messaging, so that each party of CAD cooperation can view the CAD drawing at any time during the communication and discussion, which is convenient and fast.

The utility model provides a vehicle machine, a vehicle machine system and a vehicle. The utility model relates to the technical field of vehicles. The vehicle machine comprises a micro-control unit, a tire pressure monitoring module, a power supply module, a communication module, an antenna module, a central processing unit, an information interaction module, a video module and an audio module, the micro-control unit is in communication connection with the tire pressure monitoring module; the micro-control unit is in circuit connection with the power supply module; the micro-control unit is in communication connection with the communication module; the communication module is in communication connection with the antenna module; the communication module is in communication connection with the central processing unit; the micro-control unit is in communication connection with the central processing unit; and the central processing unit is in communication connection with the information interaction module, the video module and the audio module, so that the efficiency of information transmission between the vehicle machine and the vehicle networking communication box is improved.

The present invention discloses a drinking reminder method, a water cup, and a smart device. The drinking reminder method comprises the following steps: S1: setting a drinking reminder condition; S2: calculating a user's drinking amount and drinking frequency; S3: when the drinking reminder condition is met, issuing a drinking reminder message. The technical solution of the present application can accurately measure and continuously correct an individual's daily drinking amount, and use a smart device to remind children to develop good drinking habits, thereby helping children grow up healthily and promoting their enlightenment education.

The application provides a press operation information and operation and maintenance service edge computing system and method, which is independent of a press control PLC, comprises a dual-core edge computing controller, a data acquisition module, a communication module and sensors deployed on the press, has press information acquisition and data processing, fault diagnosis, motion pair temperature and friction block life analysis and prediction, remote data management and other functions, interacts with the main control PLC in real time to exchange press information data, and uploads the data information of the press to a cloud platform to provide support for remote service of the press. The application can obtain more press state information, effectively overcomes insufficient data analysis and processing capacity of the press main control PLC, and moves the detection sensors to the edge side, shortens the main control PLC scanning cycle, realizes real-time sensing and rapid response to the press operation information state, avoids overload, overheating, excessive wear and brake failure, and improves the safety, reliability and intelligent level of the press.

The invention relates to a cooperative firepower attack intelligent decision-making system based on a CPS. The cooperative firepower attack intelligent decision-making system comprises an intelligent decision-making center, a fire control system, a weapon system, a situation awareness system, a combat command and control center, a combat data center, reconnaissance equipment and a real-time communication network. The situation awareness system is used for acquiring weapon and target information of a battlefield and transmitting the information to the intelligent decision center; the intelligent decision-making center fuses and processes information from all the systems to obtain a cooperative firepower strike decision-making scheme; the combat command control center is used for receiving the preliminary task decision and sending the final task decision to the intelligent decision center; the fire control system is used for receiving the attack decision task and controlling the weapon system to attack the target; the combat data center is used for providing data support; and the real-time communication network is used for realizing real-time information transmission. According to the method, the advantages of information-physical fusion of the decision making system are fully exerted, the cooperative firepower strike decision is optimized, and the overall firepower strike capability of the combat unit is improved.

The invention relates to a calculation method for load conversion of a local coordinate system, and the method is based on a part simulation model, and comprises the following steps: S1, determining a hard point coordinate of a part; s2, establishing a local coordinate system; s3, establishing a local coordinate load calculation formula: F < x > Local = a * F < x > Globe + b * F < y > Globe + c * F < z > Globe; s4, setting boundary conditions; s5, solving a coefficient; the method has the advantages that a model of the part does not need to be arranged during calculation, calculation can be carried out only through the hard point position, specific simulation analysis software is not needed, calculation is easy, the calculation amount is reduced, the calculation capacity is high, calculation efficiency is improved, large-scale and batch mechanical analysis calculation of the automobile part can be carried out, and the method is suitable for popularization and application. The application range is wide and the practical value is high.

The present invention provides a cloud computing-based narrow linewidth 1200nm semiconductor laser data processing system, which relates to the field of data processing technology, including: a laser emission module, which uses a semiconductor laser to emit a laser beam with a narrow linewidth of 1200nm; a data acquisition module, which is connected to the laser emission module, and obtains the state parameters of the laser beam in real time, including power, wavelength, linewidth and working environment data, and pre-processes and digitally converts the signal output by the sensor; a cloud processing platform module, which is used to receive and process the data of the data acquisition module, realize real-time analysis, anomaly detection, trend prediction and remote monitoring of the data, and the analysis results of the cloud processing platform. The present invention realizes remote intelligent monitoring and automatic optimization and adjustment of the laser, and improves the data processing efficiency and the stability of the laser performance.