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

144results about How to "Improve error" patented technology

Intelligent blood pressure measuring method and system based on pulse wave method

The invention relates to an intelligent blood pressure measuring method and system based on a pulse wave method, and belongs to the technical field of medical detection. The blood pressure measuring method comprises the steps that pulse wave signals collected by a pulse wave signal collecting device from the body surface of a subject are obtained in real time; filtering and de-noising the pulse wave signal to obtain a pre-processed pulse wave signal; extracting a plurality of characteristic parameters from the preprocessed pulse wave signals to obtain a characteristic parameter data set; analyzing the correlation between the characteristic parameters in the characteristic parameter data set and the blood pressure based on a characteristic value method, and extracting to obtain a sensitive characteristic value data set; based on a pre-established proportional relation model, calculating according to the sensitive characteristic value data set to obtain a preliminary blood pressure value; inputting the preliminary blood pressure value and the sensitive characteristic value data set into a pre-trained neural network model to obtain an optimized final blood pressure value; and outputting and displaying the final blood pressure value through a display module. The precision and stability of blood pressure measurement can be improved.
Owner:BEIJING HUAYI JINGDIAN BIOTECHNOLOGY CO LTD

Pancreatic cancer target area automatic sketching method and system

The invention discloses a pancreatic cancer target area automatic sketching method and system, belongs to the technical field of target area sketching, and solves the problem of low automatic sketching accuracy in the prior art. The method comprises the following steps: acquiring a plain-scan CT image, an enhanced CT image and a corresponding label of a pancreatic cancer patient to construct a training sample set; respectively training the constructed endangered organ segmentation model and the radiomics model based on the training sample set; training the constructed target region segmentation model based on the training sample set, the trained endangered organ segmentation model and the radiomics model to obtain a trained target region segmentation model; and obtaining an organ-at-risk segmentation result and a tumor distribution probability prediction result of the patient to be sketched based on the trained organ-at-risk segmentation model and the trained radiomics model, and inputting the enhanced CT image, the organ-at-risk segmentation result and the tumor distribution probability prediction result into the trained target region segmentation model to obtain a pancreatic cancer target region sketching result. And efficient and accurate automatic sketching of the pancreatic cancer target region is realized.
Owner:PEKING UNIVERSITY THIRD HOSPITAL (THE THIRD CLINICAL MEDICAL SCHOOL OF PEKING UNIVERSITY) +1

Transmission path construction method, interference analysis method and interference elimination method based on unmanned aerial vehicle communication and inductance integration

The invention discloses a transmission path construction method, an interference analysis method and an interference elimination method based on unmanned aerial vehicle communication and sensing integration, and the method comprises the steps: 1, building an unmanned aerial vehicle communication and sensing integration scene and a multi-input multi-output communication channel model which is composed of two unmanned aerial vehicles, a base station, a plurality of users and a moving target; step 2, establishing a communication channel model among the unmanned aerial vehicle, the base station and respective user receiving ends; and step 3, establishing an unmanned aerial vehicle sensing channel model. Mathematical formulas of all communication signals and sensing signals in an unmanned aerial vehicle communication and sensing integrated scene are defined, the functions of the unmanned aerial vehicle, a base station and a user are defined, then transmission paths of the communication signals and the sensing signals are designed, finally different interferences are analyzed under the system and the paths, and two interference elimination methods are provided. And the interference between the users, the interference between the unmanned aerial vehicles and the interference between the users and the unmanned aerial vehicles are represented by heat maps, and an interference elimination method is selected according to different interference degrees of the heat maps.
Owner:NANJING UNIV OF POSTS & TELECOMM

PIR type permanent magnet synchronous motor dead-beat prediction control method with disturbance suppression

The invention discloses a PIR-type permanent magnet synchronous motor dead-beat prediction control method with disturbance suppression, which belongs to the technical field of motor control and comprises the following steps: designing a PIR-type dead-beat current prediction controller based on an electromagnetic equation of a permanent magnet synchronous motor; designing a model reference adaptive system MRAS based on a nonlinear ADRC adaptive law; parameter identification is carried out based on MRAS; and performing disturbance observation of the PIR type deadbeat current prediction controller based on the MRAS. According to the invention, the integration link and the quasi-resonance link are added, so that the steady-state error resistance and the alternating current disturbance resistance of the dead-beat controller are improved. By adding MRAS parameter identification and disturbance observation, the anti-motor electromagnetic parameter perturbation and direct current disturbance capability of the dead-beat controller is improved. And by adopting the nonlinear ADRC adaptive law, the overshoot problem of the linear ADRC adaptive law can be improved, and the high-frequency disturbance resistance of the linear ADRC is reserved.
Owner:ANHUI NANHAO ELECTRONIC TECHNOLOGY CO LTD

Method and device for controllable load resources to participate in primary frequency modulation of power system

The invention relates to the technical field of power system automatic control, and discloses a primary frequency modulation method and device for controllable load resources to participate in a power system, and the method comprises the steps: obtaining the load state data of the power system, carrying out the multi-dimensional load clustering of the load state data, and obtaining a load feature time sequence data set; based on the load characteristic time sequence data set, establishing a Koopman operator high-dimensional linear model of the power system by using an extended dynamic mode decomposition algorithm; performing optimization solution on the load state of the power system by using the Koopman operator high-dimensional linear model of the power system to obtain an optimal load control sequence; and performing frequency adjustment on controllable load resources participating in the power system by using the optimal load control sequence. According to the method, the stable prediction of the load state of the power system and the accurate adjustment of the frequency in the controllable load resources participating in the power system are realized, and the operation efficiency and the stability of the power system are improved.
Owner:CHINA THREE GORGES CORPORATION

Fault detection method and system for charger

The invention relates to the technical field of fault detection, and discloses a fault detection method and system for a charger, and the method comprises the steps: obtaining the charging state of a battery, and the charging state of the battery comprises the charging current, the charging power, the state of charge, and the temperature of the battery; performing feature extraction on the battery charging state, and constructing a charging feature matrix; inputting the charging characteristic matrix into a battery health prediction model to obtain a battery charging demand prediction value; performing state analysis and calculation according to the battery charging state to obtain a charger state; performing correlation analysis according to the charger state and the battery charging demand prediction value to obtain fault prediction information; and obtaining a charger fault detection result according to the fault prediction information. The method can improve the fault detection efficiency of the charger.
Owner:ZHEJIANG AIFICO ELECTRIC TECHNOLOGY CO LTD

Soil moisture inversion method based on dynamic two-channel model and related device

The invention discloses a soil moisture inversion method based on a dynamic two-channel model and a related device, and the method comprises the steps: carrying out the resampling of an obtained vegetation index of an inversion time period through a space averaging method, and carrying out the reprojection of the vegetation index to an equal-area extensible earth grid, performing filtering and linear interpolation on the vegetation index time sequence of each pixel to obtain a daily dynamic vegetation index in an inversion time period; calculating the vegetation water content through the daily dynamic vegetation index, and then calculating the initial value of the vegetation optical thickness through the vegetation water content; calculating a surface roughness parameter based on a vegetation optical thickness initial value calculated based on a climate state vegetation index or a daily dynamic vegetation index; and inputting the surface roughness parameter and the vegetation optical thickness initial value into a zero-order radiation transmission model to invert the soil moisture. According to the method, the vegetation features are represented by using the daily dynamic vegetation index, and the change features of the vegetation can be described more accurately in time and space scales, so that the precision of soil moisture inversion is improved.
Owner:SUN YAT SEN UNIV

Magnetic storm identification method based on critical frequency of F2 layer of ionized layer

The invention discloses a magnetic storm identification method based on ionosphere F2 layer critical frequency. The magnetic storm identification method comprises a magnetic storm start identification method, a magnetic storm end identification method and a magnetic storm event confirmation method. The magnetic storm starting identification method comprises the following steps: extracting an F2 layer critical frequency observation value of an ionized layer vertical detection result and an expected value of the ionized layer in a calm state, and calculating the F2 layer critical frequency difference value percentage. According to the method disclosed by the invention, the magnetic storm can be effectively identified, and faults of a radio system depending on ionosphere propagation, such as short-wave communication, serious decrease of the highest available frequency of a high-frequency radar, even loss of an available working frequency band, failure of functions of the high-frequency radar, obvious increase of measurement and control and navigation positioning errors and the like, can be analyzed, researched and judged.
Owner:CHINA INST OF RADIO PROPAGATION

Smart park management platform based on three-dimensional visualization

The invention relates to the technical field of three-dimensional visualization park management, and particularly discloses a three-dimensional visualization-based smart park management platform, which comprises a self-adaptive sensor calibration module for dynamically calculating a sensor data accuracy coefficient to ensure the precision of data acquisition; the multi-dimensional data fusion and three-dimensional modeling module is used for calculating an adaptation coefficient to evaluate the response capability of the system to environmental change by integrating multi-source data; the intelligent image data registration module optimizes a matching precision coefficient and improves the space docking precision of the image data and the three-dimensional model; the real-time error monitoring and feedback optimization module uses a machine learning algorithm to correct an error between the sensor and the three-dimensional model, dynamically calculates an error correction coefficient, and ensures data consistency; and the comprehensive performance evaluation module calculates a sensor data accuracy coefficient, an adaptation coefficient, a matching precision coefficient and an error correction coefficient through normalization to obtain a comprehensive performance coefficient, and optimizes a system operation strategy in real time.
Owner:HEFEI HYZNC INTELLIGENT ENG

Adaptive multi-fidelity data fusion method based on multi-credibility model

The invention discloses a self-adaptive multi-fidelity data fusion method based on a multi-credibility model. The method comprises the following steps of: (1) inputting high and low credibility analysis sample quantities into an algorithm model, and optimizing high and low credibility space mapping parameters and training parameters; (2) calculating high and low credibility analysis model response samples, obtaining corresponding model response values, and obtaining high and low credibility sample data sets; (3) calculating the correlation between the samples with high and low credibility; (4) defining a space mapping matrix, and generating a high-reliability sample space mapping loss function and a low-reliability sample space mapping loss function; (5) performing high and low credibility spatial mapping optimization to obtain a spatial mapping matrix; (6) projecting the high credibility to the low credibility space through the space mapping matrix; step (7), establishing a multi-credibility neural network; and step (8), outputting a multi-credibility prediction result. According to the invention, self-adaptive multi-credibility data fusion is realized.
Owner:XIDIAN UNIV

Method and system for evaluating carrying capacity of power distribution network

The invention discloses a power distribution network bearing capacity evaluation method and system. The method comprises the following steps: acquiring photovoltaic output error data; inputting the photovoltaic output error data into the improved generative adversarial network for continuous training to obtain photovoltaic output scenes, clustering the photovoltaic output scenes to obtain feature scenes and feature scene probability distribution, and constructing a feature scene probability distribution uncertainty set by adopting KL divergence; based on the photovoltaic output scene and the power distribution network constraint condition, establishing a power distribution network bearing capacity evaluation model; and based on the feature scene probability distribution uncertainty set and the power distribution network bearing capacity evaluation model, establishing a power distribution network bearing capacity evaluation distribution robust optimization model, and obtaining a power distribution network bearing capacity evaluation value and a power distribution network optimal decision scheme. Errors caused by a traditional scene generation method are effectively improved; the distribution network zone area bearing capacity evaluation distribution robust optimization model constructed by using KL divergence can flexibly adjust the conservative property of the solution, and also considers the accuracy and robustness at the same time.
Owner:YUNNAN POWER GRID CO LTD

Magnetic storm identification method based on ionosphere oblique detection

The invention discloses a magnetic storm identification method based on ionospheric oblique detection. The method comprises a magnetic storm start identification method, a magnetic storm end identification method and a magnetic storm event confirmation method. The magnetic storm start identification method comprises the following steps: extracting a maximum available frequency observation value of an ionosphere oblique detection result F2 layer propagation mode and an expected value of an ionosphere in a calm state, and calculating a difference percentage. According to the method disclosed by the invention, the magnetic storm can be effectively identified, and faults of a radio system depending on ionosphere propagation, such as short-wave communication, serious decrease of the highest available frequency of a high-frequency radar, even loss of an available working frequency band, failure of functions of the high-frequency radar, obvious increase of measurement and control and navigation positioning errors and the like, can be analyzed, researched and judged.
Owner:CHINA INST OF RADIO PROPAGATION

Encoder precision optimization method, mechanical arm joint and system

The invention provides an encoder precision optimization method, a mechanical arm joint and a system.The method comprises the steps that a first encoder and a second encoder are calibrated, and encoding data are collected in real time; linear mapping is carried out on the collected coded data and actual data, and signal processing is carried out; respectively calculating a first real-time angle value of the first encoder and a second real-time angle value of the second encoder; and in response to an absolute position obtained through power-on and position and speed control during operation of a driver, obtaining numerical values of the first encoder and the second encoder, and calculating the absolute position through redundancy so as to compensate the precision of the first encoder, and electrifying to obtain the absolute position based on the first real-time angle value and the second real-time angle value.
Owner:ENCOSMART TECHNOLOGY (BEIJING) CO LTD

Whole vehicle steering function safety test evaluation method and device, vehicle and medium

The invention relates to the technical field of automobile testing and evaluation, in particular to a method and device for testing and evaluating the safety of a whole automobile steering function, an automobile and a medium. Obtaining vehicle controllability evaluation, a maximum lateral acceleration value, vehicle lateral displacement in target time, fault takeover time, steering wheel torque, a steering wheel angle and time domain curve and vehicle data information of a yaw angle and time domain curve when the test vehicle fails under the target steering test working condition; and according to a test evaluation criterion corresponding to the target steering test working condition, analyzing the multiple pieces of vehicle data information, thereby generating controllability of the test vehicle under multiple dimensions, and further generating a steering system safety evaluation report of the test vehicle. Therefore, the problems that the error of a test evaluation result is increased while the labor cost is increased and the steering system safety mechanism cannot be efficiently and accurately verified and judged when the safety performance is evaluated manually in the related technology are solved.
Owner:CATARC AUTOMOTIVE TEST CENT TIANJIN CO LTD

Electromagnetic kettle

The invention discloses an electromagnetic water-boiling kettle, comprising a base, a kettle body that can be placed on the base for matching use, an electromagnetic heater installed on the base, a main circuit board that is arranged in the base and can be electrically connected to an external power supply, and also comprising a water inlet valve arranged at the bottom of the kettle body, a water injection joint that can be lifted and moved on the base and sealably docked with the water inlet valve, a lifting device for driving the water injection joint to lift and move, and a lifting switch for controlling the lifting device; the lifting device is connected to the water injection joint and is electrically connected to the main circuit board through the lifting switch. The electromagnetic water-boiling kettle adopts a bottom water inlet method, and can still maintain good sealing performance after long-term use.
Owner:GUANGZHOU JIGU ELECTRIC APPLIANCE TECH CO LTD

Operation detection method and system of organic light-emitting device

The invention relates to the technical field of organic electroluminescence, provides an operation detection method and system of an organic electroluminescence device, and aims to solve the problem that the operation performance of the device in different states cannot be accurately detected in the prior art. Through obtaining device configuration information and a detection task, a device operation state is detected in real time, historical data is collected, a deep learning model is trained, and an operation model group corresponding to the device state is established. And adopting a time-phased detection strategy, matching optimal model processing data, outputting a detection task and optimizing a scheduling instruction. And if the error of the detection result exceeds the range, optimizing the scheduling model and early warning a fault. According to the invention, the detection accuracy is improved, the fault early warning capability is enhanced, the service life of the device is prolonged, the user experience is improved, and effective support is provided for operation and maintenance of the organic electroluminescent device.
Owner:NAT WHALE TECH (GUANGDONG HENGQIN GUANGDONG & MACAO DEEP COOP ZONE) CO LTD

Photovoltaic ultra-short-term power prediction method based on ATT-BI-LSTM

The invention relates to a photovoltaic ultra-short-term power prediction method based on ATT-BI-LSTM. The method comprises the following steps: step S1, collecting photovoltaic field station data; s2, preprocessing the data collected in the step S1; s3, performing variable selection through Kendall correlation analysis; step S4, an ATT-BI-LSTM model is constructed; and step S5, the ATT-BI-LSTM model is adopted to predict the photovoltaic ultra-short-term power. Compared with the prior art, the method has the advantages that the prediction precision is improved, meanwhile, the dimension and complexity of data can be reduced, and the calculation efficiency of the model is improved.
Owner:STATE POWER INVESTMENT GRP JIANGSU NEW ENERGY CO LTD +1

Method and system for quantitative analysis of lubricating oil additive

The invention relates to the technical field of lubricating oil analysis, in particular to a method and system for quantitative analysis of a lubricating oil additive. The method comprises the following steps: acquiring base oil data and lubricating oil additive data; performing mixed oil treatment based on the base oil data and the lubricating oil additive data to generate mixed lubricating oil additive data; performing mixed oil concentration detection on the mixed lubricating oil additive data to generate mixed oil concentration detection data; performing chromatographic column selection on the mixed lubricating oil additive data based on the mixed oil concentration detection data to generate chromatographic column selection data; and performing boiling point heating on the chromatographic column selection data to obtain mixed oil boiling point heating data. By optimizing data acquisition, chromatographic separation, component identification and quantitative analysis as well as result analysis and visualization, the accuracy of quantitative analysis of the lubricating oil additive is improved.
Owner:XIAN MARKOTE NEW MATERIAL TECH CO LTD

Nitrogen and phosphorus pollution source analysis method based on pollution load integration

The invention relates to the field of electric digital data processing, in particular to a nitrogen and phosphorus pollution source analysis method based on pollution load integration, which comprises a point source nitrogen and phosphorus pollution load accounting step, a non-point source nitrogen and phosphorus pollution load accounting step, an integrated model parameter calibration step and a nitrogen and phosphorus pollution load calculation step. And calculating the nitrogen and phosphorus pollution load of the industrial wastewater, calculating the nitrogen and phosphorus pollution load of the urban domestic wastewater, and then accounting the nitrogen and phosphorus pollution load of the point source. Calculating the dissolved nitrogen and phosphorus pollution load capacity, then calculating the adsorbed nitrogen and phosphorus pollution load capacity, and then calculating the non-point source nitrogen and phosphorus pollution load. And calculating the comprehensive nitrogen and phosphorus pollution load according to the total amount of the point source nitrogen and phosphorus pollution load and the non-point source nitrogen and phosphorus pollution load. According to the method, the point source and non-point source nitrogen and phosphorus pollution load calculation modes are integrated, and the correction RUSLE soil erosion general equation is combined to calculate the nitrogen and phosphorus pollution loads in various phase states, so that the accuracy and the reliability in nitrogen and phosphorus pollution comprehensive source analysis can be obviously improved.
Owner:TECH CENT FOR SOIL AGRI & RURAL ECOLOGY & ENVIRONMENT MINIST OF ECOLOGY & ENVIRONMENT

Mass spectrum data processing and identification method based on full-connection variational neural network

The invention relates to a mass spectrum data processing and identification method based on a full-connection variational neural network, and the method comprises the steps: S1, obtaining mass spectrum original data and pathological annotation information, and carrying out the data preprocessing; s2, inputting the preprocessed data into an unsupervised learning model based on variational self-coding, learning low-dimensional feature representation of mass spectrum data, and realizing data visualization; s3, associating the low-dimensional feature representation learned by the model with features in the original data to realize peak value learning, and obtaining a potential feature ion list related to a tumor cluster or anatomy; and S4, inputting the low-dimensional feature representation learned by the model and the potential feature ions into a Gaussian mixture model for clustering analysis, and manually selecting an optimal clustering model in combination with pathological annotation information to realize co-localization of the cluster of the region of interest and the feature ions. Compared with the prior art, the method has the advantages that efficient memory use and efficient peak value recognition are realized, and the clustering speed is higher while the accuracy is ensured.
Owner:NANCHANG ZHONGKE MOBILE MEDICAL IMAGING RESEARCH INSTITUTE

Prefabricated corrugated steel pipe spiral stirrup cage confined concrete core column reinforced composite shear wall and assembling method thereof

ActiveCN118958533AQuick Position AssemblySmall lateral stiffnessWallsProtective buildings/sheltersMesh reinforcementRebar
The invention relates to the technical field of prefabricated concrete structures, and particularly discloses a prefabricated assembly type corrugated steel pipe spiral stirrup cage confined concrete core column reinforced composite shear wall and an assembly method thereof.The composite shear wall comprises a prefabricated concrete wall and a corrugated steel pipe spiral stirrup cage compound confined high-strength concrete core column; the prefabricated concrete wall is divided into an upper-layer prefabricated concrete wall and a lower-layer prefabricated concrete wall, and a horizontal post-pouring connecting layer is arranged between the upper-layer prefabricated concrete wall and the lower-layer prefabricated concrete wall. The corrugated steel pipe spiral stirrup cage compound constraint high-strength concrete core column sequentially penetrates through the upper-layer prefabricated concrete wall, the horizontal post-pouring connecting layer and the lower-layer prefabricated concrete wall from top to bottom. A reinforcing mesh formed by interweaving vertically-distributed reinforcing steel bars, horizontally-distributed reinforcing steel bars and tie steel bars is embedded in the prefabricated concrete wall. When the upper-layer prefabricated concrete wall and the lower-layer prefabricated concrete wall are assembled, vertically-distributed steel bars in the upper-layer prefabricated concrete wall and the lower-layer prefabricated concrete wall are connected and anchored in a horizontal joint bar plug pin mode.
Owner:SOUTH CHINA UNIV OF TECH

Particle swarm optimization support vector machine algorithm for dynamically adjusting inertia weight

The invention relates to the field of artificial intelligence, in particular to a particle swarm optimization support vector machine algorithm for dynamically adjusting inertia weight, which specifically comprises the following steps: S1, initializing; s2, calculating the fitness; s3, updating individual optimization; s4, updating the global optimum; and S5, termination conditions. Through the improved PSO (WPSO) algorithm, the error penalty parameter and the kernel function parameter capable of optimizing the SVM classification effect are more efficiently determined; the improved PSO algorithm tends to fall into a local optimal solution in the later stage of the algorithm and the later-stage stability problem caused by other improved PSO algorithms, and aims to achieve better optimization performance.
Owner:梅睿彪

Large-dynamic and high-gain underwater weak laser echo signal processing method and circuit

The invention belongs to the technical field of signal processing, and discloses a large-dynamic and high-gain underwater weak laser echo signal processing method and circuit, and the circuit comprises a power conversion circuit, an avalanche photodiode, a transimpedance amplification circuit and the like. The power conversion circuit converts the input 24V direct current voltage into nominal voltage values for normal working of various circuit modules, and reduces ripple waves of the power supply to the maximum extent and reduces noise of the power supply; the transimpedance amplification circuit is used for converting a current signal output by the APD into a voltage signal; the signal extraction circuit is used for reducing the bottom of an echo signal so as to reduce the strength of a backscattered signal of a water body, retain the wave crest of the echo signal and improve the proportion of a target echo signal; the variable gain amplification circuit enables the system to have a large gain adjustable range, and the adaptability of the system to different water environments and targets is improved; and the DAC conversion circuit, the main processor and the RS232 communication module are used for adjusting the amplification gain of the circuit and the clipping quantity of the signal extraction circuit on the echo signal intensity in real time.
Owner:NAVAL UNIV OF ENG PLA

High-precision positioning method, device and system based on multi-channel fusion

The invention is suitable for the technical field of wireless communication and positioning, and provides a high-precision positioning method, device and system based on multi-channel fusion, and the method comprises the steps: obtaining the positioning data estimation of a plurality of positioning units, at least including a 5G communication positioning unit, a UWB positioning unit and a Bluetooth positioning unit; acquiring positioning data estimation reliability parameters of each positioning unit, and allocating data weights to each positioning data unit based on the reliability parameters; based on the data weight, performing weighted position fusion on the positioning data estimation of each positioning unit to obtain fused positioning estimation; and obtaining the fusion positioning estimation, and outputting the fusion positioning estimation to an application layer. According to the scheme, the data estimation of a plurality of positioning units is fused, and the complementarity of different positioning technologies can be utilized, so that the positioning precision is remarkably improved. And each unit has different expressions in different environments and scenes, and data fusion can reduce errors of a single technology, so that the accuracy of overall positioning is improved.
Owner:CITIC GUOAN INFORMATION TECH

Double-loop closed-loop control method for weak coupling resonator

The invention provides a double-loop closed-loop control method for a weakly coupled resonator, double loops comprise a first closed-loop and a second closed-loop, the first closed-loop is an amplitude regulation and control loop, the amplitude of a driving signal is adjusted by tracking the amplitude information of a single resonator in the weakly coupled resonator, and the amplitude of the driving signal is adjusted by tracking the amplitude information of the single resonator in the weakly coupled resonator. Therefore, the vibration amplitude of the specified single resonator is kept constant; and the second closed loop is a phase-locked amplifier demodulation loop and is used for tracking the phase and frequency of the other single resonator in the weakly-coupled resonator and adjusting the phase and frequency information of the driving signal so as to maintain the frequency stability of the weakly-coupled resonator. Amplitude, frequency and phase information of the driving signals obtained by the first closed loop and the second closed loop are processed by a frequency mixing link to obtain the driving signal after frequency mixing and the first driving signal. According to the invention, the weak coupling resonator can stably work in any basic resonance mode, and the resolution and precision of the sensor based on the weak coupling resonator can be improved.
Owner:AEROSPACE INFORMATION RES INST CAS

Crop yield estimation method and device based on linear time sequence characteristics and multi-scale attention

The invention provides a crop yield estimation method and device based on linear time sequence features and multi-scale attention, and the method comprises the steps: constructing a yield estimation model which comprises a linear time feature extraction module and a multi-scale attention feature module, obtaining training data formed by crop yields of a plurality of land parcels in historical years and a remote sensing image sequence corresponding to each land parcel, wherein each land parcel is labeled with a yield label; the linear time feature module regards the crop yields of the plots in the training data in historical years and the corresponding years as a time sequence, and environment mode features of the time sequence are extracted through linear regression; the multi-scale attention feature module extracts remote sensing features of a remote sensing image sequence in the training data by using attention mechanisms of different scales; the environment mode features and the remote sensing features are sent into a decoder of the multi-layer perceptron, and the predicted yield is obtained.
Owner:INST OF COMPUTING TECH CHINESE ACAD OF SCI

Deep neural network mixing precision hardware acceleration circuit based on probability calculation

The invention relates to the field of integrated circuits and artificial intelligence, in particular to a deep neural network hybrid precision hardware acceleration circuit based on probability calculation. According to the invention, operation of the deep neural network is accelerated from the level of a hardware circuit, error compensation is carried out on an existing fine-granularity space parallel probability multiplier, an error compensation probability multiplier is designed, the defect that the error of a calculation result of the existing fine-granularity space parallel probability multiplier is relatively large is overcome, and the error compensation probability multiplier is more accurate. The hybrid precision hardware acceleration circuit is used as a basic operation unit of the hybrid precision hardware acceleration circuit of the deep neural network based on probability calculation, a 3bit multiplier used in an existing hybrid precision operation processing unit is reduced to 2bit granularity, and the defects that the existing hybrid precision operation processing unit is large in area, high in delay and high in power consumption are overcome. Therefore, the probability operation of the deep neural network model is accelerated.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA

Four-cavity sliding curtain wall connection node structure and construction method

The present invention relates to a four-chamber slidable curtain wall connection node structure and a construction method thereof, the connection node structure comprising: a support frame fixedly mounted on a main structure, a limited position slideway formed on the support frame; and a curtain wall keel for connecting a curtain wall unit, a slider matched with the limited position slideway fixedly mounted at the end, the slider inserted into the limited position slideway to connect the curtain wall keel to the support frame, and the slider can be moved and adjusted along the limited position slideway, so that the end of the curtain wall keel can be moved and adjusted relative to the main structure. The present invention enables the installed unit to achieve axial displacement adjustment through the cooperation of the slider and the limited position slideway, can adjust the error of the main structure caused by construction reasons, is convenient for on-site installation, and can also ensure that the curtain wall unit and the main structure maintain the degree of freedom in the in-plane direction, avoid the curtain wall unit from being subjected to the internal force generated by the deformation of the main structure, and ensure the structural safety of the curtain wall unit.
Owner:CHINA CONSTR EIGHT ENG DIV CORP LTD

Intelligent machine tool collaborative processing system and method

The application discloses a kind of multi-intelligent machine tool collaborative machining system and method, belong to intelligent manufacturing technical field.The system includes data acquisition module, augmented reality auxiliary detection module, meta-universe platform, virtual simulation module and big data analysis module, wherein meta-universe platform receives standardized multi-dimensional processing data stream and AR interaction event, virtual simulation module calls the digital twin model of completed state synchronization to predict processing whole process data, and outputs structured simulation prediction report;Big data analysis module generates optimal adjustment strategy for machine tool parameters by parallel multi-algorithm fusion analysis, and forms collaborative scheduling instruction set.The application generates AR interaction event by augmented reality auxiliary detection module, so that processing site state data and operator's on-site confirmation, marking, adjustment and abnormal report information can jointly participate in the state synchronization of digital twin model, improve the integrity and real-time performance of multi-intelligent machine tool operation state mapping.
Owner:HUAIYIN INSTITUTE OF TECHNOLOGY