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

57results about How to "Simplify the acquisition process" patented technology

Power document intelligent question and answer method and system based on large language model

The invention belongs to the technical field of artificial intelligence, and relates to a power document intelligent question and answer method and system based on a large language model. The method comprises the steps of obtaining document data according to a preset period; preprocessing the document data to obtain an adaptive document; constructing a vector retrieval library through an embedded model according to the adaptive document; obtaining a user query question, inputting the user query question into a vector retrieval library, performing retrieval based on a vector similarity retrieval method and a keyword retrieval method to obtain related text blocks, and screening out the most related text block from the related text blocks through a two-stage rearrangement technology; a user queries the question and the most relevant text block and inputs the question and the most relevant text block into the large language model to obtain preliminary intelligent answer information; and correcting the preliminary intelligent answer information based on a preset self-provincial inversion mechanism to obtain final intelligent answer information. Multiple rounds of questions and answers and context understanding are supported based on a correction mechanism, context consistency can be kept in continuous query of complex technical questions, and the accuracy of multiple rounds of questions and responses is ensured.
Owner:DATANG HYDROPOWER SCI & TECH RES INST CO LTD +2

Zero sample image classification method and device

The invention discloses a zero sample image classification method and device, and belongs to the field of image recognition. According to the method, a category description document is generated through a large language model, global and local feature information of an image and the document is obtained through a visual encoder and a text encoder, then decoupling is performed through a visual semantic decomposition module and a text semantic decomposition module, and visual semantic embedding and text semantic embedding are generated; through joint training of the modules, optimizing total loss composed of local-to-semantic mapping variance loss, multi-semantic diversity loss, decomposition semantic alignment loss and local semantic alignment loss; and the training is used for image category prediction in the reasoning stage after completion of training. According to the invention, the difficulty of zero sample image classification can be reduced, and the classification accuracy and efficiency are improved.
Owner:INSTITUTE OF INFORMATION ENGINEERING CHINESE ACADEMY OF SCIENCES

Transformer hotspot inversion method based on time convolutional neural network, and electronic equipment

The invention discloses a transformer hotspot inversion method based on a time convolutional neural network, and electronic equipment. The method comprises the steps that a first training data set is acquired, and the first training data set is constructed based on an orthogonal design mode and a finite element simulation mode; performing data dimension reduction processing on the first training data set to obtain a target training data set; training the initial hotspot inversion model by using the target training data set, and determining a target hotspot inversion model; and using the target hot spot inversion model to predict hot spot data of the target transformer to obtain a prediction result, the prediction result including a target hot spot temperature value and a target hot spot position. According to the invention, the technical problems of low accuracy and high cost of the determination method of the hot spot temperature value and the hot spot position when the three phases of the transformer are unbalanced in the prior art are solved.
Owner:FOSHAN POWER SUPPLY BUREAU GUANGDONG POWER GRID +1

Multi-view fusion emotion recognition method under sleep deprivation condition

The invention discloses a multi-view fusion emotion recognition method under a sleep deprivation condition, and the method comprises the steps: constructing and training a recognition model which comprises a Transform network based on multi-view fusion in an offline stage, carrying out the multi-view fusion of an input signal through employing the trained recognition model in an online stage, and achieving the real-time detection. According to the method, the accuracy of emotion recognition under a single mode (EEG signal) is improved by introducing the multi-view fusion Transform model, the dependence on multi-mode data is reduced, and the influence of the sleep state on the emotion is evaluated more accurately by improving the experimental design and eliminating the interference of material fixation on the result.
Owner:SHANGHAI JIAOTONG UNIV

Clay pore pressure static sounding parametric modeling and pore pressure prediction method

The invention discloses a clay pore pressure static sounding parametric modeling and pore pressure prediction method. The method comprises the following steps: S1, establishing a two-dimensional axisymmetric CPTu model in advanced simulation software; s2, defining a soil elastic and plastic constitutive model; s3, assembling the probe and the soil body, and aligning the origin of coordinates; s4, creating an analysis step, and starting a large deformation option; s5, implementing a contact algorithm to determine a friction coefficient; s6, defining load and boundary conditions; s7, performing grid division, and optimizing a contact area; s8, compiling a parameterized script from the original script, and creating and associating an automatic calculation plug-in; and S9, extracting and calculating pore pressure data. The method simplifies pore pressure prediction, provides reliable reference for field tests, and has high prediction precision and practicability.
Owner:TSINGHUA SHENZHEN INTERNATIONAL GRADUATE SCHOOL

Method for real-time collection and manual analysis of webpage data without fixed format based on plug-in

The invention discloses a plug-in-based real-time acquisition and manual analysis method for webpage data without a fixed format. Real-time integrated operation of data acquisition, labeling and uploading is realized through a browser plug-in. A user selects target data through a floating window interface, and a plug-in automatically extracts content, identifies a platform source, performs quick labeling in combination with a preset analysis label and uploads the data to a background. According to the method, the flexibility of manual judgment and the high efficiency of an automation technology are fused, and the method can adapt to multiple platforms including high-protection dynamic webpages without developing a customized tool for a target website. The floating window supports operations such as dragging and zooming, so that the user experience is improved; the labeling is completed by data acquisition, so that the processing chain is effectively shortened, and the efficiency and accuracy are remarkably improved. The system supports authority management, encrypted transmission and real-time statistical analysis, and ensures data security and efficient utilization.
Owner:SHENZHEN SKIEER INFORMATION TECH CO LTD

Method for diagnosing deep fractured reservoir parameters based on distributed optical fiber temperature measurement data

The invention discloses a method for diagnosing parameters of a deep fractured reservoir based on distributed optical fiber temperature measurement data. The method comprises the following steps: acquiring actually measured bottom hole temperature and pressure data of the deep fractured reservoir by using a distributed optical fiber sensor; obtaining approximate analytic solutions of the bottom hole pressure at the early stage and the later stage of reservoir production; constructing a reservoir production early stage heat transfer model, and obtaining an approximate analytic solution of the production early stage bottom hole temperature; constructing a reservoir production later-stage heat transfer model, and obtaining an approximate analytical solution of the bottom hole temperature in the production later-stage; establishing a deep fractured reservoir parameter diagnosis method by using the bottom hole temperature approximate analytic solutions at the early stage and the later stage of production and the bottom hole pressure approximate analytic solutions at the later stage; and according to the actually measured bottom hole temperature and pressure data of the reservoir and a reservoir parameter diagnosis method, determining solved values of the reservoir parameters. The method provided by the invention is simple and easy to understand, has strong operability, and can improve the diagnosis efficiency of the deep fractured reservoir parameters.
Owner:SOUTHWEST PETROLEUM UNIV

Training method of self-supervised microscopic image processing neural network

The invention discloses a training method for a self-supervised microscopic image processing neural network, and the method comprises the steps: obtaining a fluorescent microscopic image data set through employing an optical microscopic imaging system with a time-dependent single photon counting function, and enabling the fluorescent microscopic image data set to comprise a photon counting distribution set {Gmn (t)} MN for M * N pixel points, m and N are respectively integers greater than 1, m is from 1 to M, n is from 1 to N, and t represents time; and constructing a training data set for a neural network comprising a fluorescence lifetime prediction convolution module and / or a denoising convolution module from the photon counting distribution set of the M * N pixel points in a photon flow redistribution mode.
Owner:INSTITUTE OF BIOPHYSICS CHINESE ACADEMY OF SCIENCES

Antibody screening method and system based on differential evolution algorithm and artificial intelligence model

The invention provides an antibody screening method and system based on a differential evolution algorithm and an artificial intelligence model. The method comprises the following steps: S1, acquiring data of an antibody-antigen compound sequence and affinity; s2, encoding the antibody and antigen sequences by using a language model, and converting the encoded antibody and antigen sequences into a matrix M; s3, training a machine learning model by using the matrix M and the affinity parameters to obtain a trained optimal machine learning model; s4, predicting the antigen affinity of the antibody by utilizing the trained optimal machine learning model; and S5, carrying out affinity optimization on the antibody sequence based on a differential evolution algorithm, and carrying out further filtration by predicting the druggability of the antibody so as to screen the antibody with high affinity and good drug properties.
Owner:SHANGHAI JIAOTONG UNIV

Method for determining horizontal ultimate bearing capacity of distributed post-grouting pile foundation

The invention is suitable for the technical field of pile foundation bearing capacity prediction, and relates to a distributed post-grouting pile foundation horizontal ultimate bearing capacity determination method, which comprises the following steps: obtaining the structure size and material parameters of a pile foundation, the physical and mechanical parameters of a soil body, the grouting amount and the grouting material parameters; establishing a pile-soil three-dimensional finite element model; selecting a soil constitutive model, and inputting the parameters into the soil constitutive model; pile-soil contact interface parameters, pile end ultimate bearing capacity standard values and pile end spring stiffness are determined; grid division is conducted on the pile-soil three-dimensional finite element model; a gravity load is applied to the pile-soil three-dimensional finite element model, and a horizontal load is applied to the pile top; boundary conditions are applied to the pile-soil three-dimensional finite element model; calculating and analyzing the pile-soil three-dimensional finite element model to obtain a P-Y curve of pile top horizontal displacement and load, a pile body stress distribution curve and a pile body internal force distribution curve; and determining the horizontal ultimate bearing capacity of the pile foundation. According to the method, the accuracy of the determined horizontal ultimate bearing capacity of the pile foundation is effectively guaranteed.
Owner:CHINA RAILWAY 16TH BUREAU GRP CO LTD +2

Method for determining ultimate uplift bearing capacity of distributed post-grouting pile foundation

The invention relates to the technical field of pile foundation uplift ultimate bearing capacity calculation, in particular to a distributed post-grouting pile foundation uplift ultimate bearing capacity determination method which is characterized by comprising the following steps that S1, parameters of a pile foundation are obtained according to pile foundation construction data; s2, a three-dimensional finite element model is established according to the pile foundation structure size in S1; s3, selecting a soil constitutive model; s4, inputting the physical and mechanical parameters of the soil in the S1 into the soil constitutive model in the S3; s5, pile-soil interface parameters are determined according to the physical and mechanical parameters, provided in the S1, of the pile foundation and the soil; s6, according to the pile foundation loading process, a finite element model calculation step is set; s7, the finite element model is calculated and analyzed, a relation curve of the pile top load Q and the drawing amount U is obtained, and the pile body axial force distribution condition is obtained through calculation; s8, according to the Q-U curve obtained through calculation in the S7 and the pile foundation uplift ultimate bearing capacity judgment condition, the distributed post-grouting pile foundation uplift ultimate bearing capacity is determined; the ultimate uplift bearing capacity of the pile foundation calculated by the method is relatively accurate.
Owner:SOUTHWEST JIAOTONG UNIV +2

Multi-modal zero sample defect detection method based on search engine and human guidance

According to the multi-modal zero-sample defect detection method based on the search engine and human guidance, the exogenous background image is automatically acquired through the search engine based on the keyword, and the acquisition process of the background image is greatly simplified; meanwhile, a textual feature extractor is enhanced by introducing a texture element network, so that the textual feature extractor can be combined with texture information in an exogenous image, accurate segmentation of the defect is realized when an unseen material or an unseen defect category is processed, and the limitation of the prior art in facing rare defects is overcome; that is to say, the problem that a traditional defect detection algorithm depends on a large-scale data set is solved by introducing a multi-modal zero sample learning method and combining feature alignment of the text and the exogenous texture image.
Owner:BEIJING INST OF TECH

Dynamic resistance strain signal acquisition device

The utility model discloses a dynamic resistance strain signal acquisition device, which comprises a detection seat, two ends of the detection seat are respectively provided with an installation part, a detection arm is arranged between the two installation parts, a detection hole is arranged in the middle of the detection arm, and four resistance strain gauges are uniformly arranged in the detection hole; a storage groove is formed in the side face of the installation part on one side, and a signal conditioning control panel is installed in the storage groove and electrically connected with the resistance strain gauges. According to the utility model, the full-bridge resistance strain gauge and the signal conditioning control board are integrated together, so that collected strain signals can be directly amplified, then converted into digital signals and directly output in a 485 form, the data collection process is further simplified, and connection and data interaction with other equipment are facilitated; the efficiency and flexibility of data acquisition are improved, and the method is more suitable for actual application scenes such as bridge monitoring.
Owner:ZHEJIANG BOYUAN ELECTRONIC TECH CO LTD

Backup task processing method and device, electronic equipment and storage medium

The invention discloses a backup task processing method and device, electronic equipment and a storage medium, and the method comprises the steps: carrying out the interface packaging of a backup task, and obtaining a job information interface corresponding to the packaged backup task; wherein the number of the backup tasks is multiple, and the multiple backup tasks are backup tasks for completing the backup process in different backup software; calling a job information interface corresponding to each backup task through the interface calling parameter to obtain task information of the backup task corresponding to each job information interface; and sending the task information of each backup task to a preset display tool, so that the task information of each backup task is displayed in a page of the preset display tool. According to the scheme, the situation that a worker needs to obtain and manage the task information of the multiple backup tasks through multiple pieces of different backup software is avoided, the error probability is reduced, the work experience of the worker is improved, and the task information obtaining efficiency is improved.
Owner:DIGITAL GUANGDONG NETWORK CONSTR CO LTD

Skin burn analysis method and system based on contrast agent blood perfusion

ActiveCN121962133AAchieve high-precision segmentationavoid blindnessImage enhancementImage analysisBurning tissueBurn tissue
The invention belongs to the technical field of burn wound analysis. According to the skin burn analysis method and system based on contrast agent blood perfusion, firstly, a fluorescence sequence after a contrast agent is injected is obtained, and a preprocessing sequence is generated through area cutting and down-sampling; a fluorescence intensity time curve of each pixel point is extracted, peak time is determined, and statistics is carried out to construct a peak time histogram; performing smooth denoising, extreme value detection and weak peak removal on the histogram, and adaptively selecting a threshold value for distinguishing normal and burnt tissues; generating an initial mask by using the threshold value, and performing connected domain analysis and morphological post-processing to obtain a target burn area mask; and finally, generating a pseudo-color image in combination with peak time. According to the method, the time dynamic characteristic of blood perfusion is fully utilized, the burn area can be accurately segmented and the microcirculation disturbance degree can be visually presented without depending on skin textures or a large amount of training data, and the objectivity and accuracy of burn depth evaluation are remarkably improved.
Owner:SHANDONG UNIV

A log download method, system, device and medium

The application discloses a log downloading method, system, device and medium, wherein the method comprises: performing space-time constraint verification on a log collection request, generating a task identifier based on a verification result, and allocating temporary storage space for the task identifier; according to topology information of a container cluster, asynchronously and in parallel, issuing a log collection instruction corresponding to the log collection request to each master node of the container cluster; based on a time window parameter in the log collection instruction, determining each target file; extracting a log compression file matched with the corresponding target file from each master node locally, and storing all the log compression files in the temporary storage space; when at least one master node is detected to complete the extraction, performing aggregated compression on all the log compression files in the temporary storage space to obtain a single log aggregation file; and downloading the single log aggregation file. The method of the application realizes stable downloading of the container cluster log.
Owner:DAWNING CLOUD COMPUTING TECH CO LTD +2

Wastewater heating, crystallizing and separating treatment device

The utility model relates to the technical field of waste water crystallization treatment, and discloses a waste water heating, crystallizing and separating treatment device which comprises a crystallizing tank, a scraper blade is connected in the crystallizing tank through a rotating assembly, an electric heating tube is fixedly connected to the inner wall of the crystallizing tank, and an electromagnetic valve is fixedly connected to the lower side of the right end of the crystallizing tank. A first support is fixedly connected to the outer surface of the lower end of the crystallizing tank, a connecting pipe is fixedly connected to the outer surface of the top end of the crystallizing tank, a filter box is fixedly connected to the upper end of the connecting pipe, a box door is slidably connected to the right end of the filter box, and a filter plate is connected to the inner wall of the filter box through a linkage assembly. The top end of the filter box is fixedly connected with a water inlet pipe. According to the device, floating objects are intercepted through the filter plate, the hydraulic rod pushes the extrusion plate to clean and extrude the floating objects, blockage is prevented, and secondary pollution is reduced; the spiral scraper scrapes crystals on the inner wall of the crystallizing tank and conveys and collects the crystals, so that the process is simplified and the cleaning trouble is avoided.
Owner:HAIZHOU E P GRP CO LTD

Power consumption test method and device, storage medium and computer equipment

The invention relates to the technical field of power consumption automatic testing, and discloses a power consumption testing method and device, a storage medium and computer equipment, and the testing method is applied to a testing host and comprises the steps that firstly, in the same wireless network, wireless communication connection with a tested terminal is established, the tested terminal is controlled to establish a target service scene, and an automatic testing action is defined; then controlling the tested terminal to execute an automatic test action in the target service scene, disconnecting the wireless communication connection with the tested terminal, then collecting power consumption data generated by the tested terminal by using the power consumption collection equipment, and finally recovering the wireless communication connection after the tested terminal completes the automatic test action. And obtaining test data generated by the tested terminal. According to the method, the test efficiency is effectively improved, the situation that the accuracy of a test result is influenced by installing a test application on the tested terminal and continuously plugging and unplugging a USB data line is avoided, and the method can be applied to a test scene with a high requirement on power consumption test precision and has very high flexibility.
Owner:BEIJING XUEDIRUANJIAN DEVELOPMENT CO LTD

Bleed air pressure evaluation method and system

The invention provides a bleed air pressure evaluation method, which is used for evaluating the bleed air pressure of a bleed air position of an intermediate casing outlet of a gas compressor, and comprises the following steps: acquiring the average total pressure and the reduced flow of the bleed air position, determining the wall surface static pressure of the bleed air position through a static pressure determination formula, and then evaluating the bleed air pressure by assisting the determined wall surface static pressure. The invention also provides a bleed air pressure evaluation system for realizing the bleed air pressure evaluation method. According to the bleed pressure evaluation method, the bleed pressure of the bleed position of the outlet of the middle casing of the gas compressor can be quickly evaluated.
Owner:AECC COMML AIRCRAFT ENGINE CO LTD

An in-vivo radio frequency response modeling method for active implantable medical devices based on equivalent lumped element model

The application discloses a kind of active implantable medical device in-vivo radio frequency response modeling method based on equivalent lumped element model, by constructing and matching the electromagnetic characteristics of the actual conductive path of AIMD and surrounding tissue, equivalent lumped element model (LEM), the key factors such as the multiscale electrical characteristics of the geometry size of implantable medical device and its in homogeneous, non-uniform even real anatomic tissue environment are parameterized characterization, so as to quickly and accurately reconstruct the radio frequency transfer function of implant.The method of the application effectively breaks through the bottleneck of the existing Tier 3 evaluation method, meets the technical requirements of the latest standard based on Tier 4 evaluation method for modeling the radio frequency transfer function of implant in the environment closely related to clinical tissue, and provides a new technical path and implementation means for efficient and accurate evaluation of the radio frequency heating risk of implantable medical device in the magnetic resonance environment.
Owner:NANCHANG UNIV

Method and device for judging initial position of slice and application of method and device

The invention provides a slice initial position judgment method and device and application of the slice initial position judgment method and device. The method comprises the steps that signals are collected through a light source and a CCD sensor, initial light intensity and a light blocking threshold value are set to judge the light blocking position, a light intensity change curve is analyzed to determine the slice position, the sample position can be calculated according to the cover glass thickness, and the curve is analyzed through an image processing algorithm. And an alarm function is provided to prompt equipment abnormity or no slice condition. The device comprises a laser projection module, a scanning platform, a concave lens, a CCD sensor and a control unit. The electronic device executes the method through a processor running program, and the readable storage medium stores related computer programs. The initial position of the slice can be simply and quickly determined, and the focusing accuracy and the working efficiency are improved.
Owner:SHENZHEN SHENGQIANG TECH

A method and system for spatial registration of visible light images and infrared images

The application discloses a kind of spatial registration method and system of visible light image and infrared image, it is related to spatial registration technical field, the position coordinate in the position coordinate in the coordinate system of second bird's-eye view obtained by visible light camera is registered with the position coordinate in the position coordinate in the coordinate system of second bird's-eye view obtained by infrared camera, more accurate conversion relationship can be obtained, and the acquisition process of conversion relationship is simpler than prior art, does not need a large number of training images, low in cost, strong practicality.
Owner:BEIJING SINOITS TECH

Converter valve side bushing on-line insulation monitoring method based on multi-dimensional time sequence characteristics

The application discloses a kind of converter valve side bushing on-line insulation monitoring method based on multidimensional timing characteristics, belong to the on-line monitoring technical field of capacitive equipment.The application obtains the mapping relationship of multiple harmonic resistance and direct current resistance by carrying out off-line test to converter valve side bushing;While synchronously collecting the leakage current and voltage signal of bushing, the mapping relationship is used to calculate timing resistance, multiple harmonic capacitance, multiple harmonic dielectric loss as insulation characteristic parameter, and then the correlation coefficient between each corresponding timing insulation characteristic parameter of the same batch of bushing is calculated;The correlation coefficient between all bushings is used to form sequence and is normalized to obtain evaluation sequence, find out the position of abnormal value and mark, while generating and updating evaluator, finally accurately locate abnormal bushing through the result of evaluator.So, the application realizes high sensitivity on-line monitoring of abnormal state caused by insulation deterioration by using mutual supervision of multiple insulation characteristic parameters between large quantities of bushings.
Owner:HUAZHONG UNIV OF SCI & TECH +1

Metadata acquisition method based on AIGC

The invention discloses a metadata acquisition method based on AIGC, and relates to the technical field of data acquisition. Comprising the steps of 1, pre-registering data sources: pre-registering and managing multiple data sources, 2, identifying and formulating an acquisition task, 3, generating task configuration and executing the task, 4, verifying a source library and generating a report: after the acquisition task is completed, performing source library verification on acquired metadata, and after the verification is passed, generating a detailed acquisition report, and step 5, carrying out manual reexamination and exception handling: carrying out manual reexamination on correct data in the acquisition report for online use, carrying out acquisition task correction on abnormal data occurring in the acquisition report, and carrying out acquisition, reporting and reexamination processes again until the data is accurate and successfully online.
Owner:INSPUR ZHUOSHU BIG DATA IND DEV CO LTD

Method and device for acquiring inquiry sheet

The invention discloses a method and a device for acquiring an inquiry sheet, and relates to the technical field of internet medical treatment. A specific embodiment of the method for obtaining the inquiry sheet comprises the following steps: in response to a received inquiry request, analyzing parameters carried by the inquiry request to obtain a heterogeneous data identifier; determining an associated heterogeneous data set according to the heterogeneous data identifier; and determining a query condition corresponding to the query request, and obtaining a target inquiry sheet meeting the query condition from the heterogeneous data set. According to the embodiment, the acquisition process of the inquiry sheet can be simplified, and the acquisition efficiency and accuracy of the inquiry sheet can be improved.
Owner:BEIJING JINGDONG TUOXIAN TECH CO LTD

Method for multi-gas measurement based on gfc wheels

PendingCN122689698ASimplify the acquisition processReduce computing requirements
The present application relates to absorption spectroscopy technology, and specifically provides a multi-gas measurement method based on GFC wheel, comprising the following steps: A1. obtaining standard deviation R of reference signal in multiple detection periods rsd and standard deviation M of measurement signal rsd ; A2. obtaining offset parameter K=f(R rsd , M rsd , t) according to the standard deviations, t is the absolute value of the difference between the rotation period at the last calibration and the current rotation period; A3. comparing the offset parameter K with a set range; if the offset parameter K is normal, no adjustment is needed; if the offset parameter K is abnormal, corresponding adjustment measures are implemented according to the set range where the offset parameter K is located, and the adjustment measures comprise: obtaining complete reference signal and measurement signal, thereby obtaining Δt, and replacing. The present application has the advantages of accurate measurement and the like.
Owner:HANGZHOU PENGPU TECH CO LTD

Power equipment information acquisition method, system, device and medium

PendingCN122248372AImplement the get methodachieve acquisitionServices signallingConnection management
This application discloses a method, system, device, and medium for acquiring information from power equipment, belonging to the field of power electronics technology. The method includes: power equipment responding to a connection request from a pre-installed device by establishing a communication connection with the pre-installed device via an AP hotspot; receiving an inquiry command sent by the pre-installed device; obtaining the response information corresponding to the inquiry command; and sending the response information back to the pre-installed device. This application, by activating the AP hotspot of the WiFi module after the power equipment is powered on, allows the pre-installed device to establish a communication connection with the power equipment via the AP hotspot, obtain the corresponding response information based on the network relationship within the power equipment cluster, and provide feedback. This eliminates the need for manual scanning of QR codes or manual data entry for each power equipment, simplifying the power equipment information collection process, shortening the information acquisition time cycle, and improving the efficiency of power equipment information acquisition.
Owner:SHANGHAI SIGE DIGITAL TECHNOLOGY CO LTD

PDAC stage prediction method and device combining deep learning and machine learning

The invention discloses a PDAC staging prediction method and device combining deep learning and machine learning. The method comprises the following steps: acquiring a serum metabolic spectrum data set; constructing different combination models randomly composed of a deep neural network, a convolutional neural network and a long short-term memory network; different hybrid models based on different combination models are constructed, the hybrid models adopt the combination models to realize feature extraction of the serum metabolism spectrum data, and classification is performed according to the extracted features by adopting a machine learning algorithm; training different hybrid models based on the data set, obtaining the trained different hybrid models, comparing the performance, and selecting the hybrid model with the best performance as a prediction model; and utilizing the prediction model to realize stage prediction of the PDAC. According to the method, different types of neural networks are constructed to perform staging prediction of different clinical stages of PDAC, and the optimal hybrid model is determined by comparing the classification performance of different clinical staging cases of PDAC by different hybrid models.
Owner:XIAMEN UNIV

A method, system, device and storage medium for detecting device anomaly

The embodiment of the present invention provides a method, system, device and storage medium for detecting device abnormalities. The method includes: within a preset time length, using a three-light acquisition device to collect data from a target device, obtaining a three-light fusion image sequence data, inputting it into an abnormal feature extraction model for feature extraction, obtaining multivariate time series sequence data, performing subsequence division on part or all of the data in the multivariate time series sequence data, obtaining multivariate subsequence segment data, inputting it into a preset abnormal detection model for error judgment, and obtaining the abnormal state judgment result of the target device. By subsequencing the multivariate time series sequence data after abnormal feature extraction, multivariate subsequence segment data reflecting device abnormalities is obtained on the basis of ensuring data accuracy, so that the present invention is improved in both detection accuracy and detection efficiency compared with the prior art. It can be seen that the present invention improves the efficiency of abnormal detection of power equipment.
Owner:STATE GRID ZHEJIANG ELECTRIC POWER CO LTD +1

Clustering-based RIS auxiliary spatial modulation system blind detection method and system

The invention discloses a clustering-based RIS auxiliary spatial modulation system blind detection method and system, and relates to the technical field of wireless communication, and the method comprises the steps: constructing an RIS auxiliary spatial modulation system, and obtaining a receiving signal of the RIS auxiliary spatial modulation system; reconstructing a received signal by constructing a matrix to obtain a channel model; performing equivalent transformation on the channel model to obtain an equivalent channel model; converting a signal detection problem into a clustering problem through an equivalent channel model; constructing an optimized K-means clustering detector based on the amplitude and phase characteristics; and iterating a clustering problem by optimizing a K-means clustering detector to realize blind detection of the signal. According to the method, the signal detection problem in the RIS auxiliary spatial modulation system is converted into the clustering problem, the amplitude and phase characteristics of the system channel are deeply excavated through the novel K-means clustering detector, the signal detection precision and efficiency are effectively improved, and dependence on channel state information is avoided.
Owner:ZHEJIANG UNIV OF SCI & TECH