Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

1122results about How to "Run fast" patented technology

Parameter optimization method and device of coupling type transformer

The invention relates to the field of transformer parameter regulation and control, in particular to a parameter optimization method and device for a coupling type transformer. The method comprises the following steps: acquiring multi-dimensional real-time working monitoring parameters and real-time working environment state parameters of the coupling type transformer; performing nonlinear dynamic analysis and global state sensing modeling on the multi-dimensional real-time working monitoring parameters so as to construct a transformer global state sensing model; performing multi-parameter combination definition on the multi-dimensional real-time working monitoring parameters, performing dynamic operation simulation based on a transformer global state sensing model, and collecting operation simulation response data of a plurality of schemes; performing multi-target dynamic balance analysis on the operation simulation response data of the plurality of schemes to obtain an optimal transformer working parameter combination; and carrying out environment dynamic parameter change analysis on the real-time working environment state parameters, and constructing a real-time environment condition distribution field. According to the invention, efficient and self-adaptive adjustment transformer parameter optimization is realized.
Owner:SHENZHEN RUIQIZHENG TECH CO LTD

Abnormal value-based weight activation quantification method after large model training

The invention discloses a weight activation quantification method based on abnormal value large model training. The weight activation quantification method comprises a weight quantification process and an activation value quantification process during reasoning. In weight quantization, model bit width is converted into low bit representation in an off-line mode, and in weight activation value quantization during reasoning, a quantized weight model is initialized, input of a linear layer is pre-scaled during reasoning operation, and finally 8-bit quantization operation is executed. According to the method, mathematical equality transformation of weight and activation is combined, a strategy of lossless quantization and low-delay execution is provided, and the significant weight is considered in an off-line weight quantization process, so that the influence of a smooth activation abnormal value on the significant weight is reduced, and the quantization error of the abnormal value is reduced in a model reasoning process. According to the method, effective quantification during large-scale model reasoning is realized, the memory required for running a large model is remarkably reduced, and the running speed is increased.
Owner:THE ACAD OF TIANJIN UNIV HEFEI

Near-electricity construction operation safety early warning method and system

The invention relates to a near-electricity construction operation safety early warning method and system, and belongs to the field of near-electricity construction operation equipment early warning, and the method comprises the steps: scanning a power transmission and transformation project construction site through three-dimensional laser, and forming a high-precision laser point cloud three-dimensional data model; quickly dividing an electrified body and a non-electrified body in the three-dimensional laser point cloud model, and carrying out attribute hooking on the safe distance range and the electrified body; rTK positioning sensors are installed on an operator and a construction machine, and high-precision real-time positioning and spatial distance measurement of the operator and the machine are achieved through Beidou RTK; real-time space distances between operators, machines and electrified bodies are monitored on the basis of a safety distance detection algorithm, reminding in different modes such as safety warning and prohibition warning is set according to requirements of safety operation regulations, and it is ensured that the operators, machines and the like keep enough safety distances from the electrified bodies.
Owner:STATE GRID SHANDONG ELECTRIC POWER CO +1

Operation inspection equipment fault detection method and system based on improved deep learning model

The invention belongs to the technical field of computer application, and relates to an operation inspection equipment fault detection method and system based on an improved deep learning model, and the method comprises the steps: obtaining original data, employing EEMD to decompose the original data for feature enhancement, obtaining a decomposition signal, carrying out the noise reduction of the decomposition signal, and obtaining a feature signal after noise reduction; according to the feature signals after noise reduction, a deep learning model is constructed based on Resnet and a dynamic Transform encoder, and the prediction probability of each fault category is output; a ZOA optimization algorithm is used to optimize hyper-parameters of a deep learning model, an optimization model is obtained, the performance of the optimization model is evaluated, an optimal model is obtained, and the prediction probability of each fault category is output, so that false report and missing report of faults can be effectively reduced in a fault detection task, potential safety hazards caused by the fact that the faults are not found in time are reduced, and the fault detection efficiency is improved. Therefore, the stability and reliability of the power system are improved.
Owner:STATE GRID BEIJING ELECTRIC POWER CO +1

Spectrum fitting method based on multiband LED spectrum and luminous flux matching

The invention discloses a spectrum fitting method based on multiband LED spectrum and luminous flux matching, which comprises the following steps of: establishing a mapping relation between a spectrum database and spectrums corresponding to different driving currents according to actual spectral characteristics of an LED, establishing a standard spectrum LTg (lambda) relational expression with luminous flux setting, taking a white light LED as a base spectrum of a fitting spectrum, and establishing a standard spectrum LTg (lambda) relational expression with luminous flux setting; and taking other monochromatic LEDs as discrete spectrums to obtain a spectrum curve needing to be compensated, performing discrete compensation on the spectrum curve needing to be compensated to obtain a driving current and a fitting spectrum when the monochromatic LED of each channel is fitted optimally, and obtaining a distribution curve and luminous flux of the fitting spectrum. According to the invention, the required target spectrum is fitted through light mixing of various monochromatic LEDs with different wavebands, the luminous flux of the fitted spectrum is matched, the control of the light intensity is realized, the method can be suitable for more illumination scenes and calibration environments, meanwhile, the calculated amount in the spectrum fitting process is reduced, the fitting precision is higher, and the method is suitable for large-scale popularization and application. And the consistency with the actual spectral characteristics of the LED device is higher.
Owner:CHANGZHOU INST OF OPTOELECTRONICS TECH +1

Water quality prediction method based on multi-source data

The invention discloses a water quality prediction method based on multi-source data, and constructs a water quality prediction model based on a long short-term memory network (LSTM) and a graph neural network (GNN) of multi-source data (surface water monitoring data, meteorological data, hydrological data and pollution source on-line monitoring data), so that short-term real-time on-line prediction of surface water quality factors can be realized. According to the method, on the basis of multi-source data such as section water quality monitoring data, historical meteorological data, forecast meteorological data, hydrological monitoring data, pollution source online monitoring data and upstream and downstream relations, multiple main water quality influence factors such as water quality, meteorology, hydrology, pollution sources and upstream and downstream section water quality are comprehensively considered; the LSTM or the graph neural network is adopted to establish a water quality prediction model, and the comprehensive influence of various influence factors and the time-space relationship on the water quality is simulated, so that the simulation result is more accurate and reliable.
Owner:CHONGQING INST OF GREEN & INTELLIGENT TECH CHINESE ACAD OF SCI +3

Key thread bundle scheduling method and device

The invention discloses a key thread bundle scheduling method and device, and relates to the technical field of GPGPU (General Purpose Graphics Processing Unit), a thread bundle which is executed in sequence and a thread bundle which is branched are scheduled at different frequencies, different scheduling schemes are appointed for the two types of thread bundles in the mode, the parallel performance of the sequence thread bundle is ensured not to be influenced to the greatest extent, and the scheduling efficiency is improved. And meanwhile, the operation of the branch thread bundles is accelerated, the problem that the operation time of the thread bundles is prolonged due to excessive thread branches is avoided, and the combining performance of the universal graphics processor is improved to a certain extent.
Owner:SHANDONG INSPUR SCI RES INST CO LTD

Smart city video monitoring adjustment method and system based on 5G network

The invention discloses a smart city video monitoring adjustment method and system based on a 5G network, and relates to the field of smart cities, and the method comprises the steps: 1, drawing a city map through a geographic information system technology, deploying a high-resolution monitoring camera supporting the 5G network, and an edge computing device, a video stream in an H.265 compressed format is captured in real time by using a built-in image sensor. The high-resolution camera is combined with high-speed transmission of a 5G network, the real-time performance of monitoring data is ensured, emergencies are quickly responded, real-time processing of video data and sensor data is achieved, the high-resolution monitoring camera is accessed through a Video Capture class of OpenCV, target detection is conducted on an area of interest in combination with a YOLO algorithm, and the real-time processing of the video data and the sensor data is achieved. The target identification accuracy and processing efficiency are improved, the edge computing device is used for carrying out data summary on the segmented frame area, the processing delay is dynamically adjusted according to the real-time network and the computing load, and efficient operation of the monitoring system under different conditions is ensured.
Owner:SMIC (HARBIN) CO LTD

Active equalization control method for SOC of multiple battery modules of energy storage system SCLBC

The invention belongs to the technical field of power electronics and the technical field of new energy storage, and relates to an energy storage system SCLBC multi-battery module SOC active balance control method which comprises the steps that a charging process control method and a discharging process control method are adopted, sampling equipment including a BMS sends information to a controller, the controller synthesizes information such as SOC of all modules, the information is analyzed and processed, and the SOC of all the modules is obtained. The obtained single-module power regulation information is transmitted to each corresponding module, and a bidirectional DC-DC converter in each module regulates the duty ratio to make a response, so that the power reaches a set value, and a control target is completed; by monitoring the state of the battery in real time, analyzing the proper working state and dynamically adjusting the working mode of each battery to adjust the working state, the service life of the battery is prolonged, the energy consumption is ensured to be the lowest as far as possible and the balance control is completely realized, so that the problems of limited conditions and imperfect control methods of the control method in the prior art are solved.
Owner:XI AN JIAOTONG UNIV

Multi-target fusion cow oestrus behavior detection method and system

The invention discloses a multi-target fusion dairy cow oestrus behavior detection method and system, and belongs to the technical field of target detection of deep learning. The method comprises the following steps: S1, collecting a dairy cow oestrus behavior image, and performing image enhancement on the dairy cow oestrus behavior image; s2, constructing a PBF-YOLO deep learning network model, wherein the PBF-YOLO deep learning network model comprises a lightweight module, a background information enhancement module and a feature aggregation module; and S3, identifying a dairy cow oestrus behavior region of interest through the PBF-YOLO deep learning network model, and detecting the dairy cow oestrus behavior. The invention provides a solution for the situation that the existing lightweight algorithm is low in detection precision and poor in robustness due to the fact that the cow oestrus behavior and other background features are similar and are difficult to distinguish in an actual pasture and cows with different scales appear on the same image.
Owner:INNER MONGOLIA UNIV OF TECH

System and method for predictive maintenance function of switch cabinet based on thermodynamics

The invention relates to a thermodynamics-based system for a predictive maintenance function of a switch cabinet, and the system comprises a sensor unit which comprises a temperature measurement sensor and a current sensor and is used for collecting a temperature signal, a current signal and an environment temperature signal of a connection point of a main loop of the switch cabinet in real time; the alarm unit is used for sending out an alarm signal after a fault is predicted to occur when the switch cabinet is in non-full-load operation; and a control unit configured to: calculate a difference between the temperature collected from the sensor unit and an ambient temperature to obtain a temperature rise value; initializing a model based on the corresponding time, the collected current and the calculated temperature rise value to obtain five fitted parameters; performing temperature rise value prediction based on the collected current signals in the prediction time period by using the obtained five parameters; and when the switch cabinet is judged to have a fault according to the comparison result of the predicted temperature rise value and the collected measured value, the alarm unit is controlled to send out an alarm signal.
Owner:SCHNEIDER ELECTRIC (CHINA) CO LTD

Method and system for monitoring service life of water electrolysis hydrogen production system

The invention discloses a water electrolysis hydrogen production system life monitoring method and system, and the method comprises the steps: collecting the monitoring data of a target part in a system, and collecting the voltage information of an electrolytic cell in the system; calculating the loss rate of the target component according to the monitoring data of the target component; according to the loss rate of the target component and the voltage information of the electrolytic cell, calculating operation data of the target component at the end of life of the system; and controlling the current of the electrolytic bath according to the operation data of the target component at the end of life of the system and the priority mode of the system. Thus, full life cycle monitoring and life prediction are performed on the water electrolysis hydrogen production system, the service life and the safety coefficient of the water electrolysis hydrogen production system are improved, and safe operation of the system is promoted.
Owner:JIANGSU HYDROGEN CORE POWER TECHNOLOGY CO LTD

Remote sensing target detection method and system based on improved RT-DETR

The invention discloses a remote sensing target detection method and system based on improved RT-DETR. The method comprises the following steps: step 1, collecting remote sensing image data; step 2, carrying out preprocessing such as denoising, contrast enhancement and definition enhancement on the acquired remote sensing image data; 3, making the preprocessed pictures into a remote sensing data set, and dividing the remote sensing data set into a training set and a verification set in proportion; 4, training the remote sensing data set by adopting an RT-DETR model to obtain a remote sensing target detector; step 5, optimizing various parameters of the target detector on the verification set by adopting a hyper-parameter search verification mode; and step 6, detecting a test image or video by using the remote sensing target detector optimized in the step 5. Compared with an original RT-DETR model, on the premise that original precision is kept, the calculation amount is reduced, and the network convergence speed is increased.
Owner:ZHEJIANG SCI-TECH UNIV

Space target lengthened astronomical image astronomical positioning method of large-view-field space-based monitoring platform

The invention discloses an astronomical positioning method for a space target elongated astronomical image of a large-field-of-view space-based monitoring platform, relates to the technical field of astronomical observation, and solves the problem that the elongated astronomical image is abandoned due to incapability of target association and precise orbit determination caused by inaccurate centering of the elongated astronomical image and low precision of a positioning method of an existing monitoring platform. According to the method, an observation image is divided into 16 sub-view fields, and each sub-view field is an independent view field and is calculated by an independent processor; removing fixed stars in the observation image by adopting an adjacent frame image difference method and extracting a space target star image; secondly, describing the length and angle of the lengthened astronomical image by adopting a minimum enclosing rectangle, and constructing a degradation model; performing motion blurred image restoration on the elongated star image; error points with large errors are eliminated by adopting a mutual comparison method, and track association of the space target is completed; and finally completing precise astronomical positioning of the elongated astronomical image and the like. According to the method, the processing efficiency is greatly improved, and a large number of lengthened astronomical images which are abandoned before are fully utilized again.
Owner:CHANGCHUN SATELLITE OBSERVATORY OF NAT ASTRONOMICAL OBSERVATORY OF CHINESE ACAD OF SCI

Multi-center inherited metabolic disease information integration platform

The invention discloses a multi-center inherited metabolic disease information integration platform, and relates to the technical field of medical treatment, according to the multi-center inherited metabolic disease information integration platform, the normalization and availability of data are improved through a data standardization module; data from different centers are integrated into a unified database through the data integration module by utilizing a data fusion technology, redundant data are removed, the problem of heterogeneous data integration is effectively solved, and the integrity and consistency of the data are improved; the privacy protection module avoids data leakage and unauthorized access; through the resource calculation management module, the resource utilization efficiency is effectively improved, and the calculation cost is reduced; and secondly, the storage resources are dynamically adjusted and allocated according to the evaluation result of the storage index Cczs, so that the optimal management of the storage resources is realized, the problems of resource waste and resource insufficiency are avoided, and the efficient operation of the system is ensured.
Owner:GENERAL HOSPITAL OF PLA

Polygon mark point rapid identification and positioning method and system, and storage medium

The invention relates to the technical field of mark point identification and positioning, in particular to a polygon mark point rapid identification and positioning method and system and a storage medium. Then, the gradient direction information and the coordinate information of the reserved strong edge are used for rapidly estimating the rotation angle and the center of the polygon mark point to be detected, the method is small in calculation amount and easy to operate, and rough pose information of the polygon mark point to be detected can be rapidly obtained; the estimated angle is further corrected in a small range in a rotary projection mode, the deviation between an estimated value and a real value is reduced, and a more accurate initial value is provided for subsequent registration; finally, the corrected coarse positioning result is used as an initial value for registration, the convergence speed can be increased, the number of iterations can be reduced, meanwhile, the probability that iteration falls into a local extreme value can be reduced, the operation speed is increased, and the accuracy rate is increased.
Owner:HEFEI ANXIN PRECISION TECH CO LTD

Tensor serialization method and related product

The invention discloses a tensor serialization method and a related product, and generally relates to the technical field of artificial intelligence. The method comprises the following steps: responding to a view-class operator subgraph with a to-be-processed tensor, and obtaining an original tensor of the view-class operator subgraph; and according to the original tensor, carrying out serialization on the tensor to be processed. According to the method, the device, the chip, the system and the computer readable storage medium provided by the disclosure, the to-be-processed tensor has the view type operator subgraph in response to the to-be-processed tensor, the view type operator subgraph can be utilized to restore the to-be-processed tensor into the original tensor, and then continuous processing is performed based on the original tensor. Furthermore, in some embodiments, a high-performance computing library kernel operator corresponding to the view-class operator subgraph is used for carrying out data handling, so that the continuity of the tensor to be processed is realized at a relatively high speed, and the running speed of the neural network model is further improved.
Owner:CAMBRIAN (KUNSHAN) INFORMATION TECH CO LTD

Zone area photovoltaic data anomaly identification method and system

The invention relates to a transformer area photovoltaic data anomaly identification method and system, belongs to the technical field of photovoltaic power generation, and solves the problem that the existing photovoltaic data anomaly identification and anomaly attribution are inaccurate. Comprising the following steps: collecting and preprocessing transformer area photovoltaic power generation data and meteorological data, and obtaining time point data in a to-be-analyzed time period; transmitting the data of each time point into the power generation power prediction model of the transformer area to obtain a power generation power prediction value of each time point; constructing a data point pair of each time point according to the power generation power predicted value of each time point and an actual power generation power value of the same time point in the transformer area photovoltaic power generation data; obtaining abnormal point pairs from the data point pairs of the time points by using a clustering algorithm; and according to the time point of the abnormal point pair, obtaining corresponding transformer area business data, and according to a preset photovoltaic business abnormity rule, obtaining a rule category corresponding to the transformer area business data as an abnormity reason of the corresponding abnormal point pair. And efficient and accurate anomaly identification and attribution are realized.
Owner:STATE GRID SHANGHAI MUNICIPAL ELECTRIC POWER CO

Dynamic flexible job shop scheduling method considering multiple types of events

The invention discloses a dynamic flexible job shop scheduling method considering multiple types of events, which realizes simulation of shop scheduling and design of rules through establishment of a model, completes identification of dynamic events, and realizes scheduling of the multiple types of events when a single dynamic event occurs by means of a deep reinforcement learning model. Recording the occurrence number of the dynamic event scheduling rules of the type, counting a scheduling rule set with the highest occurrence frequency, and recording the scheduling rule sets with the highest occurrence frequency corresponding to different dynamic events in the same manner; when a corresponding dynamic event occurs, the corresponding action space set is used for production scheduling; by recording the tardiness, images of the tardiness and the delivery period and the completion time of each iteration are recorded, and the minimum tardiness is obtained through a tardiness iteration graph. Compared with a traditional heuristic-based scheduling strategy, the scheduling capability of the dynamic flexible job space is effectively improved, the tardiness is reduced, the operation complexity is effectively reduced, and the operation speed is improved.
Owner:ZHEJIANG UNIV OF TECH

Lightweight crack penetration degree detection method based on ground penetrating radar and Swin Transform optimization

The invention discloses a lightweight crack penetration degree detection method based on ground penetrating radar and Swin Transform optimization, and the method comprises the following steps: 1, collecting ground penetrating radar data, and carrying out the preprocessing and feature extraction; step 2, constructing a multi-feature fusion data set; step 3, designing a two-stage through crack detection model based on Swin Transform optimization; 4, a first-stage through crack detection model optimized on the basis of Swin Transformer-YOLOv8 is designed, and a second-stage through crack detection model is designed on the basis of Swin Transformer-YOLOv8 optimization; and step 5, training and predicting the data of the training set and the test set according to the model architecture in the step 3 and the step 4. According to the invention, the accuracy of crack penetration degree detection is improved, and a scientific basis is provided for road maintenance; the detection speed is greatly improved, and the requirement of large-area rapid detection is met; the calculation complexity is reduced, the equipment performance pressure is reduced, and the detection efficiency is improved; the generalization ability of the model is enhanced, and the method can adapt to different road conditions; and the development of an intelligent road detection technology is promoted.
Owner:CHANGAN UNIV

Safety belt detection method and system based on novel YOLOv10

The invention provides a novel YOLOv10-based safety belt detection method and system, and the method comprises the steps: obtaining image data, and marking the image data, namely marking the position / wearing state of a safety belt; constructing a detection model based on a novel YOLOv10 network, namely replacing partial standard convolution Conv of the Backbone network with local attention regularization convolution; using a separable convolution star module to replace an original feature extraction module of the Backbone network; a CARAFE module is used for replacing up-sampling operation in a Neck network, multi-scale space attention dilation convolution fusion modules are additionally arranged before three sets of feature maps output by the Neck network, and then the feature maps are output to a Head network; and training a detection model by using the labeled image data to obtain a safety belt detection model. When the novel YOLOv10 network is used for safety belt detection, the detection efficiency and performance can be remarkably improved.
Owner:HUNAN UNIV

Preparation process of high-thermal-conductivity composite substrate

The invention discloses a preparation process of a high-thermal-conductivity composite substrate, and the process comprises the steps: fixing a pre-prepared skeleton on an insulating base layer, and enabling branches of the skeleton and the surface of the insulating base layer to jointly define a plurality of heat dissipation spaces; dividing the plurality of heat dissipation spaces into at least one high heat dissipation space and at least one low heat dissipation space according to the layout and heating characteristics of a predetermined electronic component; a first heat-conducting filler is injected into the high-heat-dissipation space, a second heat-conducting filler is injected into the low-heat-dissipation space, curing treatment is conducted, the heat conductivity coefficient of the first heat-conducting filler is larger than that of the second heat-conducting filler, and the first heat-conducting filler and the second heat-conducting filler are combined with the framework to form a heat-conducting layer; a functional layer of the printed circuit board is fixed on the surface of the heat conduction layer, the area, corresponding to the high-heat-dissipation space, of the functional layer is used for installing high-heat-dissipation electronic elements, and the area, corresponding to the low-heat-dissipation space, of the functional layer is used for installing low-heat-dissipation electronic elements According to the technical scheme, different heat conduction measures can be adopted for different heating elements.
Owner:DONGGUAN HETONG ELECTRONICS CO LTD

SLAM dynamic disturbance suppression method based on fuzzy processing and target detection

The invention relates to the technical field of computer vision and deep learning, and provides an SLAM (Simultaneous Localization and Mapping) dynamic disturbance suppression method based on fuzzy processing and target detection, which comprises the following steps of: obtaining a fuzzy remover; performing blurring processing on the image by using a blurring remover, and extracting ORB feature points; obtaining a dynamic target detector; detecting the clear image through a dynamic target detector to obtain a target detection frame; classifying motion states of objects in the clear image according to the target detection frame and the ORB feature points, and obtaining a static feature point set by using an overlapping frame strategy and epipolar geometric constraint; and performing system pose estimation and trajectory prediction according to the static feature point set. According to the method, the adverse effect of the dynamic target on other parts of the system is reduced through deblurring, the dynamic target can be accurately recognized through the target detection function, comprehensive elimination of the dynamic feature points is achieved in combination with the overlapping frame strategy and the epipolar geometric constraint, and the overall performance of the system is well improved.
Owner:NANKAI UNIV

Oil pumping unit frequency converter dynamic regulation and control device and method based on deep learning

PendingCN119496439ASolving Speeding ProblemsReduce useless power consumptionElectric motor controlGeneral control strategiesFrequency changerPhase currents
The invention relates to the technical field of oil pumping unit control, and provides an oil pumping unit frequency converter dynamic regulation and control device and method based on deep learning, and the device comprises a motor operation data detection device which is used for the motor operation state of an oil pumping unit, and comprises a phase voltage, a phase current and a power factor; the main control unit is configured to predict a voltage value of a frequency converter of the oil pumping unit based on an aggregation model of score and weight parameters and a deep learning model according to the detected load and the motor operation state of the oil pumping unit, and send the voltage value to the full-power adaptive regulator; and the full-power self-adaptive regulator is used for connecting the main control unit and the frequency converter, is arranged on the output side of the frequency converter, and ensures that the motor of the oil pumping unit runs in the optimal interval by self-adaptively regulating and controlling the regulation and control instruction of the frequency converter according to the received predicted voltage value. The device disclosed by the invention can realize intelligent tracking of the motor along with load change, and can realize harmonic suppression and energy feedback.
Owner:SHANDONG HOTEAM ELECTRICAL +1

Microneedle and treatment method thereof

The invention relates to a microneedle and a treatment method thereof, which realize automatic identification and accurate positioning of human acupuncture points by integrating components such as a high-precision sensor array, a controller, an output module and a user interface. Wherein the sensor array can detect electrical characteristics, tissue density change and temperature change under skin, and assists in identifying acupuncture point positions; an energy column array is arranged in the output module, pulse waves with specific frequency and energy are released, energy flow is simulated, deep stimulation is carried out on acupuncture points, and therefore the treatment effect is enhanced. In addition, the system is provided with a user interface, so that treatment parameter setting, treatment mode selection and treatment record viewing are facilitated. The method comprises the steps of system starting, automatic acupoint searching, treatment mode selecting, pulse stimulation, parameter adjusting, treatment information recording and the like, and intelligence and automation of acupuncture treatment are achieved.
Owner:GUOYIKANG TECH GRP CO LTD

Method and device for accelerating homomorphic encryption algorithm by using fast number theory transformation and medium

The invention discloses a method and device for accelerating a homomorphic encryption algorithm by using fast number theory transformation and a medium. The method comprises the following steps: initializing homomorphic encryption parameters and generating a public key, a private key and a relinearization key; encrypting a plaintext to be operated by using the public key to generate a corresponding ciphertext polynomial; performing fast number theory transformation on the ciphertext polynomial to transform the ciphertext polynomial into a point value form; performing homomorphic multiplication on the two point value form ciphertexts to obtain three ciphertexts; using a relinearization key to transform and decompose one ciphertext by using a fast number theory and accumulate the decomposed ciphertext into the other two ciphertexts to obtain two similar primary ciphertexts; using fast number theory inverse transformation to transform the first-order ciphertext from a point value form into a polynomial form; and decrypting the ciphertext polynomial by using the private key to obtain a plaintext multiplication result. The method adopts fast number theory transformation and inverse transformation, improves the polynomial multiplication calculation efficiency through the processes of primitive root sampling, bit inverse sequence and the like, is suitable for the field of polynomial operation, and improves the polynomial operation efficiency.
Owner:RENMIN UNIVERSITY OF CHINA

Universal Transform acceleration method and system based on floating point instruction set extension

The invention provides a universal Transform acceleration method and system based on floating point instruction set extension, and the method comprises the steps: constructing a multiplier which comprises a plurality of stages of adders and a 4-Booth coding algorithm which are connected in sequence, and coding data; the coded data is operated through a plurality of stages of adders in sequence to obtain the output of a multiplier; the construction of the floating-point number register comprises the following specific steps: adding a floating-point number register group, generating the floating-point number register group, enabling input result register indexes to correspond to respective register numbers by writing port logic of the floating-point number register group, generating a writing enable signal, writing data into the register by an enabled general register, and writing the data into the floating-point number register group. The read port is realized by using a multi-path parallel selector, and a selection signal of the multi-path parallel selector is an index of a read operand register; and the RISC-V processor accelerates the operation of the Transform neural network based on the constructed multiplier and the floating-point number register.
Owner:SHANDONG UNIV

Workpiece trajectory planning method and device and storage medium

The invention discloses a workpiece trajectory planning method and device and a storage medium, and belongs to the technical field of workpiece measurement. Through spline interpolation and application of a Frenet coordinate system, the method can guarantee the stable speed, achieves the optimal trajectory in time, improves the measurement efficiency and precision of the curved surface of a complex workpiece, solves the problems of many measurement points and long time consumption in a conventional method, effectively processes singular points and curvature changes on the curved surface, and improves the measurement precision of the curved surface of the complex workpiece. Therefore, the measurement precision of the scanning measuring head is higher, and the operation speed is faster. Finally, a high-precision measurement task can be completed in a short time.
Owner:TZTEK TECHNOLOGY CO LTD

Deep learning noise suppression reconstruction method applied to quantum image sensor

The invention discloses a deep learning noise suppression reconstruction method applied to a quantum image sensor, and relates to the field of image processing, and the method comprises the steps: firstly, carrying out the calculation of a block and an image of a three-dimensional frame plane sequence, and facilitating the improvement of the suppression capability of Poisson noise through precise unbiased variance stabilization transformation; secondly, the method is based on a network framework fusing Transform and CNN, the framework serves as a core denoising device and can deeply search key information in image denoising in a linear and nonlinear component combination mode, and the adaptability of the network to the denoising capacity of various scenes is enhanced through the sub-network parallel technology; and finally, using accurate unbiased variance to stabilize inverse transformation, using maximum likelihood estimation to restore a real photon number, and starting from a QIS imaging principle, further suppressing noise in a reconstructed image. According to the method provided by the invention, the reconstructed image noise of the QIS can be remarkably suppressed, and the method has the potential of being applied to a customized neural network processor chip, so that the method can obtain a relatively high operation speed.
Owner:THE ACAD OF TIANJIN UNIV HEFEI

Voiceprint-based 10kV circuit breaker mechanical fault detection method

The invention discloses a voiceprint-based 10 kV circuit breaker mechanical fault detection method. Comprising the following steps: S1, building a MobileNetYOLOv4 neural network model; s2, acquiring voiceprints through a high-sensitivity sensor; s3, performing convolution operation on a convolution kernel of edge side identification early warning; s4, circuit breaker fault identification; s5, carrying out improved S transformation on the signal; according to the invention, task allocation is dynamically calculated by a resource allocation module embedded in a cloud platform, a lightweight model is constructed, the model operation speed is accelerated, and a real-time target identification and classification system is constructed; the Moblienet neural network structure reduces the running memory of a neural network model and accelerates the running speed of the neural network model, and in order to improve the target recognition speed and solidify the model to an edge device, the MobileNet network structure applies a deep separable convolution technology, so that the target recognition efficiency is improved. The standard convolution of the convolutional neural network is divided into two steps of deep convolution and point-by-point convolution in the calculation process, so that the calculation amount of the network is greatly reduced.
Owner:NANJING YOUNENGTE ELECTRIC POWER TECH DEV