159 results about "DataOperation" patented technology

The embodiment of the invention discloses a data processing device, a chip product, computer equipment and a data processing method. The data processing device comprises an instruction decoding module, a data moving module and a plurality of calculation units; wherein the instruction decoding module is used for receiving a calculation instruction based on open source instruction set architecture RISC-V extension; the data moving module is used for responding to the calculation instruction and determining a prediction configuration parameter to be used by a target calculation unit corresponding to the calculation instruction; the calculation unit is used for utilizing a target calculation unit corresponding to the calculation instruction to obtain configuration parameters to be used by the target calculation unit when the calculation instruction is executed; and if the configuration parameter hits the predicted configuration parameter, performing unit configuration by using the predicted configuration parameter so as to perform data operation processing.

The invention discloses a non-inductive cooperative regulation and control method and device for flexible loads in a building and a medium. The method relates to the field of power control, and solves the problem that a high-precision and quick-response flexible load regulation and control scheme is lacked. According to the method, the problem of space-time correlation adaptation of the equipment is solved by collecting the running state, environment and personnel distribution multi-source heterogeneous real-time data of the equipment, constructing the topological dynamic graph of the equipment, outputting nonlinear space-time correlation information and capturing the dynamic thermal-mechanical coupling relation between the equipment in real time. A personalized regulation and control instruction is generated by an independent decision-making unit to replace a fixed threshold value; coupling disturbance is predicted in combination with associated information, initial PID parameters are compensated, it is ensured that the temperature adjusting amplitude is converged to a user perception threshold value, and frequent starting and stopping of equipment are avoided. The independent decision-making unit and PID compensation are both based on local data operation, and the instruction response period is shortened; differentiated control instructions are converted according to equipment sensitivity, a high-sensitivity area is finely regulated and controlled to ensure comfort, and a low-sensitivity area preferentially responds to power grid peak clipping.

The invention relates to the field of low-altitude anti-null data processing, in particular to a large-data-volume data processing method based on multi-level cache. According to the invention, each functional module is independently bound with a CPU core, after the previous functional module finishes processing, result data is stored in a cache and a current writing position is recorded, and the next functional module regularly reads the buffered data and records a current reading position; each function module independently uses the bound CPU to carry out data operation, and the processing delays among the function modules are not superposed, so that the data processing real-time performance is improved.

The embodiment of the invention discloses an electric leakage detection method, device and equipment, a medium and a program product. Comprising the steps of obtaining a to-be-processed signal in an electric leakage judgment period; according to the to-be-processed signal, waveform peak points in the electric leakage determination period and peak left edge points corresponding to the waveform peak points are determined; determining every two adjacent peak left edge points as a peak left edge point group, and counting the total sampling number and the trough sampling number corresponding to each peak left edge point group; according to each total sampling number, each trough sampling number, a preset correction wave period sampling number lower limit value, a preset correction wave period sampling number upper limit value and a preset trough sampling number threshold value, determining a leakage waveform from correction waveforms corresponding to each peak left edge point group; and determining an electric leakage detection result according to the distribution condition of each electric leakage waveform in the electric leakage judgment period. The complexity and the data operand of a circuit required by electric leakage detection are reduced, and the stability and the universality of electric leakage detection are improved.

The embodiment of the invention provides a storage scheduling method, a processing unit, a system on chip, computing equipment, a computer storage medium and a computer program product. The storage scheduling method is applied to a processing unit, the processing unit comprises a plurality of kernels, each kernel executes data operation of a task flow of an operation type of the kernel by adopting multi-level storage configuration of the processing unit, and the method comprises the following steps: determining estimated operation duration of each operation task in the task flow of each kernel, obtaining an estimated operation duration of each operation task of the plurality of kernels; according to the estimated operation duration of each operation task of the plurality of kernels, determining each to-be-scheduled task of each task flow of the plurality of kernels in the current scheduling time period; and scheduling each to-be-scheduled task into the multi-level storage configuration, so that a previous operation task belonging to the same task flow as the to-be-scheduled task is operated when the access duration lower limit of the storage unit scheduled by each to-be-scheduled task is reached.

1. Name of the product in this design: Edge Computing Server (RL-3000H4 4U). 2. Purpose of this design: It is used for data computing and processing, and can have functions such as computing, storage and AI acceleration. It is suitable for smart cities, industrial internet, smart retail, smart manufacturing, vehicle networking, cloud computing, 5G and other scenarios. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the design's key points: a 3D model. 5. The bottom surface of the product is a part that is not easily seen or cannot be seen during use, so the bottom view is omitted; it is already shown in the perspective view, so the top view is omitted.

The embodiment of the invention discloses a privacy intersection method and device, electronic equipment and a storage medium. The method comprises the following steps: receiving first privacy intersection data sent by a first participant; receiving second to-be-privately-exchanged data fed back by the first participant; processing the second to-be-used privacy intersection data based on the second parameter factor to obtain to-be-used privacy intersection data corresponding to the current participant; and based on the first privacy intersection data and the to-be-used privacy intersection data, determining target common data shared by the current participant and the first participant. According to the technical scheme, the problems of complicated data operation process, poor universality and low accuracy of the privacy intersection method based on the elliptic curve are solved, the data operation amount of privacy intersection is reduced, the universality of the privacy intersection method based on the elliptic curve is enhanced, and the privacy intersection efficiency and accuracy are improved.

The invention relates to a multi-expansibility intelligent service unit based on an S5000C + 310P architecture, and belongs to the technical field of computers. The intelligent service unit comprises a VPX case, an S5000C general-purpose computing module, a 310P intelligent module, a network switching module, a power module, a timing system module and a bus bottom plate, wherein the S5000C general-purpose computing module and the 310P intelligent module are interconnected through PCIe (Peripheral Component Interconnect Express). According to the method, the S5000C general computing power and the 310P AI computing power are fused, so that multi-task parallel processing is realized, the data operation and AI reasoning efficiency is remarkably improved, the processing delay is reduced, and the intensive computing scene requirements are met; through the modular integrated design, disassembly, assembly, debugging and later expansion and upgrade are facilitated, the operation and maintenance workload and cost are reduced, and the deployment flexibility is improved.

The application provides a graph data processing method, system, device and medium, comprising processing a plurality of to-be-processed graph data in a training data set to obtain node features and edge features; generating a node feature matrix and a resistance variation array according to the node features and the edge features; generating first transition features according to a graph neural network, the node feature matrix and the resistance variation array; inputting the first transition features into one or more graph convolution layers in the graph neural network for weighted calculation to output second transition features; inputting the second transition features into a readout network previously constructed in the graph neural network to generate result features; adjusting weight parameters of the graph neural network according to the result features and a Hamiltonian loss function, realizing training of the graph neural network to obtain a target neural network; and generating target result features according to node features, edge features and the target graph neural network of target graph data. The graph data operation time is reduced, and the processing speed is accelerated.

The application discloses a handshake system, method and device of a three-dimensional convolutional neural network, computer equipment and a medium. The method comprises the following steps: obtaining to-be-transported data, and writing the to-be-transported data into a cache as a cache data block in a preset size; writing the size of the cache data block into a register group according to a preset storage mode to obtain data block record information; when a data operation request is received, obtaining required data information from the data operation request, wherein the required data information comprises three-dimensional coordinates; determining whether the cache data block contains data corresponding to the required data information based on the three-dimensional coordinates and the data block record information; and if the cache data block contains the data corresponding to the required data information, generating a response signal indicating that there is valid data in the cache data block. The handshake efficiency is improved by using the application.

The application relates to a multi-stage flow water multi-path data operation and access control system, which comprises a multi-stage flow water processor group, a basic parameter configuration module, a storage-computation-access parameter configuration module, a storage-computation-access timing control training unit, a storage-computation-access scheduling controller, a CPU unit, an SOC bus, a parallel data moving module and a DDR storage unit; under the multi-stage flow water pipeline processing of multi-path input data, the data flow water processing training of storage, computation and access is carried out, the architecture configuration scheme of the multi-stage flow water processor group or processor core and the storage-computation-access timing control method of the multi-stage flow water processing are obtained through the training data operation, and the storage-computation-access processing of the multi-stage flow water is completed according to the architecture configuration scheme and the flow water processing control method obtained through the training; the system can dynamically adapt to different performances and customer demands in various application scenarios, reduces the number of SOC buses and memory units, improves the flow water processing access efficiency, improves the access and computation parallel degree efficiency, simplifies system design and saves cost.

The embodiment of the invention discloses a multi-data-type multiplier and a method for realizing multiplication of multiple types of data, relates to the technical field of digital integrated circuits, and is convenient for realizing the multiplier supporting data operation of multiple data types at lower cost, so that the area of the integrated circuit and power consumption resources are saved. The method comprises the following steps: basic multiplication units of at least two data types, wherein the basic multiplication units are used for realizing mantissa part or fixed-point number multiplication of floating-point numbers of predetermined data types through combinational operation; the control unit is used for selecting basic multiplication unit combinations of different data types according to the data types of the input matrix; the input matrix is decomposed into a plurality of parts, and the parts are respectively transmitted to corresponding basic multiplication units for operation; and performing shift and accumulation operation on the operation result of each part. The method is suitable for an integrated circuit design scene.

This invention discloses an accelerated convolutional neural network operation system, computing device, and computer storage medium. The operation system includes a first static random access memory (SRAM), a global-local buffer, and a cluster of operation units. The first SRAM stores the data required for the entire convolution operation. The global-local buffer stores frequently accessed data during the convolution operation. The cluster of operation units includes multiple operation units for performing multiplication and accumulation operations. The global-local buffer includes: an input feature map buffer for storing input feature map data and providing data sequentially according to a predefined loop order; a weight buffer, which includes a second SRAM, a feature map temporary storage area, and a first loop engine; and a partial sum buffer for storing partial summation results, including a third SRAM and a second loop engine. This invention is applicable to convolutional neural networks.

The present invention relates to sports training and testing technology and discloses a water training integrated machine for dragon boat racing. The system comprises a dragon boat, a shore data processing device, and a water device. The water device is mounted on a blade of the dragon boat and includes a sensor, a data operation unit, and a communication transmitting module, sequentially connected. These components are powered by a power supply module. The data operation unit communicates with the shore data processing device via the communication transmitting module. The shore data processing device comprises a data processing module, a display module, and multiple communication receiving modules. The display module and communication receiving modules are electrically connected to the data processing module. This technical solution enables accurate measurement and real-time monitoring of stroke force during water sports training, thereby enhancing performance analysis and training efficiency.

The invention provides an alarm method for a chemical central control system and a related device, and belongs to the technical field of alarm. According to the method, operation intervention is carried out on a DCS computer according to an alarm prompt, the DCS computer transmits an operation intervention instruction to a DCS database, and the DCS database carries out data operation processing on the operation intervention instruction and then generates a control adjustment signal; adjusting the real-time chemical production operation data according to the control adjustment signal; and the adjusted real-time chemical production operation data is fed back to the DCS database again, and the DCS database carries out data operation processing on the adjusted real-time chemical production operation data again until the adjusted real-time chemical production operation data is recovered to the preset value of the important alarm in the DCS database. The problems that the importance level of alarm cannot be distinguished, and an operator cannot be effectively prompted to focus on key alarm information are solved.

The application belongs to the technical field of image recognition, and particularly relates to an inter-frame prediction method, device and equipment based on advanced motion vector expression, which comprises the following steps: constructing a basic motion information list for a current block by using enhanced duplicate checking, wherein the basic motion information list comprises at least one basic motion information; determining a plurality of offset motion vectors by using an offset distance list and an offset direction list; offsetting a basic motion vector in each basic motion information by using the plurality of offset motion vectors to obtain a plurality of offset motion information; and calculating an offset prediction value of the current block by using the plurality of offset motion information. The method of the application can improve the accuracy of duplicate checking, reduce the repetition of basic motion information caused by omission, reduce the data operation amount, and improve the efficiency by using enhanced duplicate checking mode to check the basic motion information.

The application relates to the technical field of Java virtual machines, and particularly discloses a method for improving the performance of a JVM based on SIMD, which comprises the following steps: obtaining a loop logic instruction in Java code, processing the compiled loop logic instruction to obtain a SIMD instruction set; parsing a loop logic operation in a Java instruction to obtain an assembly instruction; sending the assembly instruction to a CPU through the SIMD instruction set; and obtaining a calculation result returned by the CPU after SIMD operation. The application parses common data loop logic operation into a SIMD instruction set, and all data are loaded into the CPU at a time in the mode of executing one operation based on the SIMD instruction set, so that the data operation efficiency is improved, the CPU is prevented from repeatedly executing simple work, the processing performance of the CPU is saved, and the running efficiency of the Java virtual machine is improved.

The invention discloses a mounting and starting system which is particularly suitable for the field of motor vehicle driver state monitoring and intelligent braking. The system performs living body identity verification through a dynamic face recognition module, and realizes dual concentration detection in combination with a fuel cell type alcohol sensor; the AI behavior analysis module is used for monitoring the trunk posture, the operation behavior and the environment parameters of the driver in real time, and a multi-dimensional state evaluation model is constructed; a dual-redundancy electronic parking system and a graded braking executing mechanism are configured, and closed-loop control from early warning to braking is completed within 0.8 second. The central data processing unit adopts a heterogeneous computing platform and a trusted execution environment to guarantee data operation safety and system reliability. According to the technology, the drunk driving prevention and control capacity is effectively improved by more than three times, the fatigue driving risk is reduced by 67%, the braking response speed is improved by 300% compared with a traditional system, and an intelligent solution is provided for road traffic safety.

The invention discloses a blanking amount correction method based on idling detection, and relates to the technical field of feeding equipment, the method comprises the following steps: S1, receiving a planned blanking amount M through a control panel, the control panel integrating data operation and instruction sending functions; s2, the theoretical pulse number P plan is calculated according to the blanking weight K of each circle and the Hall pulse number P of each circle pre-stored by the control panel, and P plan = (M / K) * P; s3, the control panel drives a driving motor of discharging equipment to start discharging, the discharging process is monitored through a state detection sensor which is installed on a material inevitable path at the bottom of a discharging pipeline, and Hall pulses generated only in the period when the state detection sensor judges that the state of materials exists are accumulated to serve as the effective pulse number P actual; s4, the control panel calculates the actual discharging amount M actual according to the effective pulse number P actual, wherein M actual = K * (P actual / P); and S5, the control panel actually transmits the actual discharging amount M to a display unit to be output.

This application discloses a data stream processing method, apparatus, device, and storage medium based on a reconfigurable TPU, including: acquiring the data to be processed and the current operating status parameters of each processing subunit; evaluating the processing complexity of the data to be processed to obtain a complexity evaluation value; determining the real-time splitting granularity corresponding to the operating status parameters of each processing subunit; adjusting the real-time splitting granularity based on the complexity evaluation value; splitting the data to be processed based on the adjusted target splitting granularity; marking the data source of the multiple data blocks to be processed obtained from the splitting; matching the operating status parameters corresponding to each processing subunit with the source identifiers corresponding to each data source mark to determine the data blocks to be processed corresponding to each processing subunit, which are then used by each processing subunit to perform data operation processing on the corresponding data blocks to be processed. This improves the accuracy and rationality of data splitting, data processing efficiency, resource utilization, and system adaptability.

The invention relates to the technical field of electronic equipment, and discloses a dual-port RAM prototype verification method and device and a prototype verification system.The dual-port RAM prototype verification device comprises a dual-port RAM module, a first port control unit, a second port control unit and a read data output control unit. An operation sequence of old data reading, operation and new data writing is completed in an external clock period, and an out-bit enabling function is simulated by using a common dual-port RAM; and meanwhile, the read data output control unit aligns, registers and outputs high-frequency operation results to ensure that the time sequence is correct. According to the scheme, the problem that the bit-enabled dual-port RAM cannot be directly supported in FPGA prototype verification is solved, and functional equivalent simulation can be efficiently and accurately realized without special hardware.

An arithmetic device (10) includes multiple PEs (12) arranged in two-dimensional manner. Each PE (12) has a selector (24) that outputs input data in a set direction. The PE (12) performs data movement via the selector (24) with another non-adjacent PE (12) located diagonally, or performs data movement via the selector (24) in own PE (12) by moving data from a register (22) to an arithmetic circuit (20), by processing a selector setting command and an input output command in a single operation.

The application discloses a kind of cloud ERP security data cross-domain processing method and system, different multi-party secure computation protocols are designed for the characteristics of different data operation performance requirements in cloud ERP, different data encryption and processing methods are used for different cloud ERP data, under the premise of guaranteeing data security and personal privacy, the fastest encryption scheme suitable for application scenario is used. Both meet the needs of users for cloud ERP data operation speed, and guarantee the security of data, realize the combination of multi-party secure computation and cloud ERP cross-domain, each computing node can be realized without exposing own privacy data, while meeting the security and reliability of data cross-domain, guarantee the data transmission speed, have higher use value and popularization value in the field of cloud ERP data cross-domain processing.

This invention provides a smart garment for capturing fitness movements and reconstructing body postures, comprising a wearable device body, a data collection system, and a data processing system. The wearable device body includes a tight-fitting sportswear, triboelectric strain sensing fibers, flexible conductive circuitry, and a packaging shell. The data collection system includes a data acquisition chip and a power module. The data processing system includes a display terminal that executes a comprehensive algorithm, which includes a deep learning algorithm, a multi-dimensional matching calibration algorithm, a feature extraction algorithm, and a body posture reconstruction algorithm. The smart garment can achieve precise full-space fitness movement data capture and body posture reconstruction. By collecting motion signals during fitness activities, identifying the name of the fitness activity, calculating joint bending angle data and comparing it with standard data, the garment can display the movement posture configuration in real time in a three-dimensional visualization simulation environment and provide real-time fitness feedback suggestions.

The present application relates to the field of computer vision and artificial intelligence, and specifically to the processing of the pooling layer of the convolutional neural network, and proposes a method and device for large-core pooling. The method mainly includes the HBLK / WBLK block mode of mapping the pooling output to the input, including: performing the operation of the pooling core kh / kw length characteristic data on the H / W direction dimension, saving the temporary results of the direction in the internal cache SRAM, storing the temporary data in the maximum output mode, and when the internal cache size range is exceeded, the H / W direction block of cblk is performed; the W and H directions represent the width and height directions of the feature map; kw / kh are the sizes of the W / H direction pooling core respectively. The device includes a pooling top layer control device, an input / output control device, an HBLK unit control device, an HBLK unit control device, and a pooling operation device. The present application can realize the operation of the pooling core with unlimited size, the hardware device proposed can achieve the balance of performance, bandwidth, power consumption and area, and can effectively solve the technical problems such as small pooling core size and high redundancy of tensor characteristic data operation in the prior art.

This invention provides a data processing unit, relating to the field of high-performance computing technology. The data processing unit includes a first-precision data masking processing module, a first-precision arithmetic module, and a second-precision arithmetic module. The first-precision data masking processing module is connected to both the first-precision arithmetic module and the second-precision arithmetic module. The first-precision data masking processing module includes a logic operation processing unit and a first multiplexer, with the logic operation processing unit connected to the first multiplexer. The data processing unit provided in this embodiment of the invention reduces module power consumption by using a forced mask to prevent abrupt changes in the input data of the first-precision arithmetic module.

The application discloses a kind of multi-sensor fusion's ship berthing intelligent monitoring system and method, including perception layer, transmission layer, processing layer and application layer, perception layer is in the bow of each berth, stern position, each deployment a set of multi-sensor integrated unit, responsible for collecting the real-time position data of ship bow and stern, transmission layer will the original data collected by perception layer be transmitted to processing center at high speed, reliably and safely, processing layer is as the core of system, is deployed on the server of data center, is responsible for all data operation, analysis and decision-making, application layer is as the interface of system and user interaction, the intelligent result of processing layer is presented in intuitive form, and control instruction is executed, the application can ensure the reliability and accuracy of system under all weather conditions, completely eliminate safety monitoring blind area, enhance the intellectualization, universality and ease of use of system, reduce manpower cost, improve work efficiency, application prospect is wide, it is favorable to popularization and application.

The application belongs to the technical field of computer vision, and particularly relates to a multi-target tracking method; on the basis of a classic Kalman filter + Hungarian algorithm, a unique PoDe-SORT framework is proposed, element-wise fusion of a width-height ratio representing the posture of a target with a standard IoU, construction of a deep iterative layered correlation process from fine granularity to coarse granularity based on pseudo-depth, and the like, so that a lightweight PoDe-SORT based on 2D sensor data operation with smaller calculation amount is realized while the robustness is improved. Overall, the application can realize the effect of target tracking based on 3D sensor information with small calculation amount, and has high robustness, strong operation rationality, small deformation error and high scene applicability.

The invention provides a training method, an application method, a training device and an application device of a parameter model, relates to the technical field of data processing, and can solve the problem that the accuracy and efficiency of manual risk slow release capability evaluation are too low. According to the specific technical scheme, the method comprises the steps of obtaining object attribute features of a plurality of sample objects; calculating a first value parameter of the sample object according to a data operation rule related to the financial supervision rule; and taking the object attribute feature and the first value parameter as training samples, training the first parameter model to obtain a trained second parameter model, and obtaining a second value parameter of the target object based on the object attribute feature of the target object so as to obtain the risk slow release value of the target object under the financial supervision rule. By implementing the scheme, the accuracy and the automation degree of risk slow release value calculation can be improved.